modelId
stringlengths 5
139
| author
stringlengths 2
42
| last_modified
timestamp[us, tz=UTC]date 2020-02-15 11:33:14
2025-08-08 18:27:49
| downloads
int64 0
223M
| likes
int64 0
11.7k
| library_name
stringclasses 495
values | tags
listlengths 1
4.05k
| pipeline_tag
stringclasses 55
values | createdAt
timestamp[us, tz=UTC]date 2022-03-02 23:29:04
2025-08-08 18:27:48
| card
stringlengths 11
1.01M
|
|---|---|---|---|---|---|---|---|---|---|
opria123/whisper-tiny-minds14
|
opria123
| 2025-04-29T03:11:34Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"whisper",
"automatic-speech-recognition",
"generated_from_trainer",
"dataset:PolyAI/minds14",
"base_model:openai/whisper-tiny",
"base_model:finetune:openai/whisper-tiny",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-04-27T12:35:56Z |
---
library_name: transformers
license: apache-2.0
base_model: openai/whisper-tiny
tags:
- generated_from_trainer
datasets:
- PolyAI/minds14
metrics:
- wer
model-index:
- name: opria123/whisper-tiny-minds14-finetuned
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: PolyAI/minds14
type: PolyAI/minds14
config: en-US
split: train
args: en-US
metrics:
- name: Wer
type: wer
value: 0.33436150524367675
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# opria123/whisper-tiny-minds14-finetuned
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the PolyAI/minds14 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8674
- Wer: 0.3344
- Wer Ortho: 0.3270
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Wer Ortho |
|:-------------:|:--------:|:----:|:---------------:|:------:|:---------:|
| 0.0005 | 34.4912 | 1000 | 0.7591 | 0.3368 | 0.3245 |
| 0.0001 | 68.9825 | 2000 | 0.8217 | 0.3307 | 0.3202 |
| 0.0001 | 103.4561 | 3000 | 0.8547 | 0.3325 | 0.3239 |
| 0.0001 | 137.9474 | 4000 | 0.8674 | 0.3344 | 0.3270 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0+cu126
- Datasets 3.5.1
- Tokenizers 0.21.1
|
zacCMU/mistral-MedQA-finetune
|
zacCMU
| 2025-04-29T03:06:03Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-04-26T22:53:49Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a π€ transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
kostiantynk1205/9d355f79-f5ab-46fb-be6a-576df0b2b87c
|
kostiantynk1205
| 2025-04-29T02:57:38Z | 0 | 0 |
peft
|
[
"peft",
"generated_from_trainer",
"base_model:tiiuae/falcon-rw-1b",
"base_model:adapter:tiiuae/falcon-rw-1b",
"region:us"
] | null | 2025-04-29T02:57:15Z |
---
library_name: peft
tags:
- generated_from_trainer
base_model: tiiuae/falcon-rw-1b
model-index:
- name: kostiantynk1205/9d355f79-f5ab-46fb-be6a-576df0b2b87c
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# kostiantynk1205/9d355f79-f5ab-46fb-be6a-576df0b2b87c
This model was trained from scratch on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.1915
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.3
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3
|
toshiya373/distilbert-base-uncased-finetuned-fake-or-real_03
|
toshiya373
| 2025-04-29T02:52:47Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"distilbert",
"text-classification",
"generated_from_trainer",
"base_model:distilbert/distilbert-base-uncased",
"base_model:finetune:distilbert/distilbert-base-uncased",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2025-04-29T02:25:57Z |
---
library_name: transformers
license: apache-2.0
base_model: distilbert/distilbert-base-uncased
tags:
- generated_from_trainer
model-index:
- name: distilbert-base-uncased-finetuned-fake-or-real_03
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# distilbert-base-uncased-finetuned-fake-or-real_03
This model is a fine-tuned version of [distilbert/distilbert-base-uncased](https://huggingface.co/distilbert/distilbert-base-uncased) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0496
- F1 Score: 0.9889
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
### Training results
### Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1
|
open-lab-taiwan/Qwen2.5-7B-Open-R1-Distill-v1-0317
|
open-lab-taiwan
| 2025-04-29T02:50:51Z | 4 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"dataset:open-r1/OpenR1-Math-220k",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-03-19T09:54:51Z |
---
base_model: Qwen/Qwen2.5-7B-Instruct
datasets: open-r1/OpenR1-Math-220k
library_name: transformers
tags:
- generated_from_trainer
- open-r1
licence: license
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
---
# Model Card for None
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the [open-r1/OpenR1-Math-220k](https://huggingface.co/datasets/open-r1/OpenR1-Math-220k) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="None", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/edward7777777sas-ntut-edu-tw/huggingface/runs/4qgk1zsw)
This model was trained with SFT.
### Framework versions
- TRL: 0.16.0.dev0
- Transformers: 4.50.0.dev0
- Pytorch: 2.5.1
- Datasets: 3.3.2
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
open-lab-taiwan/Qwen2.5-1.5B-Open-R1-Distill-v1-0317
|
open-lab-taiwan
| 2025-04-29T02:50:40Z | 3 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"dataset:open-r1/OpenR1-Math-220k",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-1.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-03-18T01:59:19Z |
---
base_model: Qwen/Qwen2.5-1.5B-Instruct
datasets: open-r1/OpenR1-Math-220k
library_name: transformers
tags:
- generated_from_trainer
- open-r1
licence: license
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
---
# Model Card for None
This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the [open-r1/OpenR1-Math-220k](https://huggingface.co/datasets/open-r1/OpenR1-Math-220k) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="None", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/edward7777777sas-ntut-edu-tw/huggingface/runs/rnk40uv8)
This model was trained with SFT.
### Framework versions
- TRL: 0.16.0.dev0
- Transformers: 4.50.0.dev0
- Pytorch: 2.5.1
- Datasets: 3.3.2
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
maksf8486/c2fb190c-1d1f-432e-88cb-3b31caf94fba
|
maksf8486
| 2025-04-29T02:47:55Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"mistral",
"axolotl",
"generated_from_trainer",
"base_model:jhflow/mistral7b-lora-multi-turn-v2",
"base_model:adapter:jhflow/mistral7b-lora-multi-turn-v2",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-29T01:48:24Z |
---
library_name: peft
base_model: jhflow/mistral7b-lora-multi-turn-v2
tags:
- axolotl
- generated_from_trainer
model-index:
- name: c2fb190c-1d1f-432e-88cb-3b31caf94fba
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: jhflow/mistral7b-lora-multi-turn-v2
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- 5676b37f940d59a0_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/5676b37f940d59a0_train_data.json
type:
field_instruction: question
field_output: response
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
dpo:
beta: 0.1
enabled: true
group_by_length: false
rank_loss: false
reference_model: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: maksf8486/c2fb190c-1d1f-432e-88cb-3b31caf94fba
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/5676b37f940d59a0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 77f3624b-a86b-48c1-ac39-c4b3682b1961
wandb_project: s56-2
wandb_run: your_name
wandb_runid: 77f3624b-a86b-48c1-ac39-c4b3682b1961
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# c2fb190c-1d1f-432e-88cb-3b31caf94fba
This model is a fine-tuned version of [jhflow/mistral7b-lora-multi-turn-v2](https://huggingface.co/jhflow/mistral7b-lora-multi-turn-v2) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.1269
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.152 | 0.0169 | 200 | 1.1269 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
liweiweigg55634/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-slithering_pesty_ferret
|
liweiweigg55634
| 2025-04-29T02:44:36Z | 4 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am slithering pesty ferret",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-22T10:21:06Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-slithering_pesty_ferret
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am slithering pesty ferret
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-slithering_pesty_ferret
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="liweiweigg55634/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-slithering_pesty_ferret", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
ahmedch28/mistral_7b_finetuned_pr_v6
|
ahmedch28
| 2025-04-29T02:44:12Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-04-29T02:44:05Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a π€ transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
cherryDavid/Qwen3-1.7B-Q4_K_M-GGUF
|
cherryDavid
| 2025-04-29T02:35:02Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"base_model:Qwen/Qwen3-1.7B",
"base_model:quantized:Qwen/Qwen3-1.7B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-04-29T02:34:54Z |
---
base_model: Qwen/Qwen3-1.7B
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-1.7B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# cherryDavid/Qwen3-1.7B-Q4_K_M-GGUF
This model was converted to GGUF format from [`Qwen/Qwen3-1.7B`](https://huggingface.co/Qwen/Qwen3-1.7B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen3-1.7B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo cherryDavid/Qwen3-1.7B-Q4_K_M-GGUF --hf-file qwen3-1.7b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo cherryDavid/Qwen3-1.7B-Q4_K_M-GGUF --hf-file qwen3-1.7b-q4_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo cherryDavid/Qwen3-1.7B-Q4_K_M-GGUF --hf-file qwen3-1.7b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo cherryDavid/Qwen3-1.7B-Q4_K_M-GGUF --hf-file qwen3-1.7b-q4_k_m.gguf -c 2048
```
|
pcam-interpretability/dino-vits16-val08398-vit-tuned-safe
|
pcam-interpretability
| 2025-04-29T02:26:49Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-29T02:26:44Z |
# dino-vits16
**Best Validation Accuracy:** `0.8398`
## Metadata
- **Model Name**: `dino-vits16`
- **Optimizer**: `adamw`
- **Scheduler**: `cosine`
- **Weight Decay**: `0.001`
- **Warmup Epochs**: `5`
- **Patience**: `10`
- **Amp**: `True`
- **Seed**: `42`
- **Batch Size**: `352`
- **Initial Lr**: `0.001`
- **Total Epochs Ran**: `38`
- **Early Stopped**: `True`
- **Training Time Seconds**: `22297.275145292282`
- **Num Parameters**: `21666049`
- **Device**: `NVIDIA A100-SXM4-40GB`
- **Run Id**: `vit-tuned-safe`
## Training Configuration
- Epochs: `38`
- Batch size: `352`
- Learning rate (initial): `0.001`
## Training Logs (Per Epoch)
| Epoch | Train Loss | Train Acc | Val Loss | Val Acc | LR |
|-------|------------|-----------|----------|---------|----|
| 1 | 0.6211 | 0.7119 | 0.4845 | 0.7484 | 0.000200 |
| 2 | 0.4682 | 0.7774 | 0.5087 | 0.7466 | 0.000400 |
| 3 | 0.4455 | 0.7918 | 0.4857 | 0.7465 | 0.000600 |
| 4 | 0.4391 | 0.7945 | 0.4634 | 0.7683 | 0.000800 |
| 5 | 0.4309 | 0.7989 | 0.4192 | 0.7904 | 0.001000 |
| 6 | 0.4169 | 0.8078 | 0.4200 | 0.7889 | 0.001000 |
| 7 | 0.3996 | 0.8167 | 0.3879 | 0.8149 | 0.000999 |
| 8 | 0.3911 | 0.8222 | 0.4468 | 0.7795 | 0.000996 |
| 9 | 0.3830 | 0.8264 | 0.3944 | 0.8103 | 0.000991 |
| 10 | 0.3759 | 0.8305 | 0.4086 | 0.7977 | 0.000984 |
| 11 | 0.3678 | 0.8353 | 0.4321 | 0.7857 | 0.000976 |
| 12 | 0.3621 | 0.8377 | 0.4163 | 0.8022 | 0.000965 |
| 13 | 0.3574 | 0.8411 | 0.3871 | 0.8170 | 0.000952 |
| 14 | 0.3543 | 0.8418 | 0.5018 | 0.7661 | 0.000938 |
| 15 | 0.3490 | 0.8453 | 0.4141 | 0.8099 | 0.000922 |
| 16 | 0.3412 | 0.8499 | 0.3623 | 0.8295 | 0.000905 |
| 17 | 0.3336 | 0.8534 | 0.4005 | 0.8193 | 0.000885 |
| 18 | 0.3565 | 0.8418 | 0.3622 | 0.8311 | 0.000864 |
| 19 | 0.3494 | 0.8447 | 0.3668 | 0.8304 | 0.000842 |
| 20 | 0.3359 | 0.8521 | 0.4189 | 0.8000 | 0.000819 |
| 21 | 0.3355 | 0.8519 | 0.3609 | 0.8314 | 0.000794 |
| 22 | 0.3280 | 0.8566 | 0.3757 | 0.8241 | 0.000768 |
| 23 | 0.3203 | 0.8606 | 0.3917 | 0.8174 | 0.000741 |
| 24 | 0.3278 | 0.8571 | 0.3974 | 0.8180 | 0.000713 |
| 25 | 0.3109 | 0.8649 | 0.3669 | 0.8310 | 0.000684 |
| 26 | 0.3129 | 0.8649 | 0.3511 | 0.8390 | 0.000655 |
| 27 | 0.3080 | 0.8667 | 0.3574 | 0.8373 | 0.000624 |
| 28 | 0.3050 | 0.8686 | 0.3584 | 0.8398 | 0.000594 |
| 29 | nan | 0.8688 | nan | 0.8383 | 0.000563 |
| 30 | nan | 0.6657 | nan | 0.5005 | 0.000531 |
| 31 | nan | 0.5000 | nan | 0.5005 | 0.000500 |
| 32 | nan | 0.5000 | nan | 0.5005 | 0.000469 |
| 33 | nan | 0.5000 | nan | 0.5005 | 0.000437 |
| 34 | nan | 0.5000 | nan | 0.5005 | 0.000406 |
| 35 | nan | 0.5000 | nan | 0.5005 | 0.000376 |
| 36 | nan | 0.5000 | nan | 0.5005 | 0.000345 |
| 37 | nan | 0.5000 | nan | 0.5005 | 0.000316 |
| 38 | nan | 0.5000 | nan | 0.5005 | 0.000287 |
|
JoseSC23/almoxcontrol
|
JoseSC23
| 2025-04-29T02:22:45Z | 0 | 0 | null |
[
"license:other",
"region:us"
] | null | 2025-04-29T02:22:44Z |
---
license: other
license_name: almoxcontrol
license_link: LICENSE
---
|
mountaingiles/mountaingile
|
mountaingiles
| 2025-04-29T02:22:16Z | 0 | 0 | null |
[
"license:creativeml-openrail-m",
"region:us"
] | null | 2025-04-29T02:22:16Z |
---
license: creativeml-openrail-m
---
|
John6666/illumiyume-xl-illustrious-v30-sdxl
|
John6666
| 2025-04-29T02:21:07Z | 0 | 0 |
diffusers
|
[
"diffusers",
"safetensors",
"text-to-image",
"stable-diffusion",
"stable-diffusion-xl",
"anime",
"girls",
"characters",
"cute",
"style",
"stability",
"trained",
"e-pred",
"Illustrious XL v1.0",
"illustrious",
"en",
"dataset:deepghs/danbooru2024",
"dataset:deepghs/e621_newest",
"base_model:OnomaAIResearch/Illustrious-XL-v1.0",
"base_model:finetune:OnomaAIResearch/Illustrious-XL-v1.0",
"license:other",
"autotrain_compatible",
"endpoints_compatible",
"diffusers:StableDiffusionXLPipeline",
"region:us"
] |
text-to-image
| 2025-04-29T02:13:52Z |
---
license: other
license_name: faipl-1.0-sd
license_link: https://freedevproject.org/faipl-1.0-sd/
language:
- en
library_name: diffusers
pipeline_tag: text-to-image
tags:
- text-to-image
- stable-diffusion
- stable-diffusion-xl
- anime
- girls
- characters
- cute
- style
- stability
- trained
- e-pred
- Illustrious XL v1.0
- illustrious
base_model: OnomaAIResearch/Illustrious-XL-v1.0
datasets:
- deepghs/danbooru2024
- deepghs/e621_newest
---
Original model is [here](https://civitai.com/models/1308285?modelVersionId=1720089).
This model created by [duongve13112002](https://civitai.com/user/duongve13112002).
|
Docty/dreambooth-marine-growth-lora
|
Docty
| 2025-04-29T02:17:35Z | 0 | 0 |
diffusers
|
[
"diffusers",
"tensorboard",
"text-to-image",
"lora",
"diffusers-training",
"stable-diffusion",
"stable-diffusion-diffusers",
"base_model:stable-diffusion-v1-5/stable-diffusion-v1-5",
"base_model:adapter:stable-diffusion-v1-5/stable-diffusion-v1-5",
"license:creativeml-openrail-m",
"region:us"
] |
text-to-image
| 2025-04-29T02:03:46Z |
---
base_model: stable-diffusion-v1-5/stable-diffusion-v1-5
library_name: diffusers
license: creativeml-openrail-m
inference: true
instance_prompt: Cinematic under water view of an object whose surface is surrounded
with severe sks growth of marine and corrosion.
tags:
- text-to-image
- diffusers
- lora
- diffusers-training
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
- diffusers
- lora
- diffusers-training
- stable-diffusion
- stable-diffusion-diffusers
---
<!-- This model card has been generated automatically according to the information the training script had access to. You
should probably proofread and complete it, then remove this comment. -->
# LoRA DreamBooth - Docty/dreambooth-marine-growth-lora
These are LoRA adaption weights for stable-diffusion-v1-5/stable-diffusion-v1-5. The weights were trained on Cinematic under water view of an object whose surface is surrounded with severe sks growth of marine and corrosion. using [DreamBooth](https://dreambooth.github.io/). You can find some example images in the following.
LoRA for the text encoder was enabled: False.
## Intended uses & limitations
#### How to use
```python
# TODO: add an example code snippet for running this diffusion pipeline
```
#### Limitations and bias
[TODO: provide examples of latent issues and potential remediations]
## Training details
[TODO: describe the data used to train the model]
|
rafsanfalle/sdvfdv
|
rafsanfalle
| 2025-04-29T02:17:26Z | 0 | 0 | null |
[
"license:bsd-2-clause",
"region:us"
] | null | 2025-04-29T02:17:23Z |
---
license: bsd-2-clause
---
|
ibrahimkettaneh/Hammer2.0-0.5b-4.5bpw-h8-exl2
|
ibrahimkettaneh
| 2025-04-29T02:16:37Z | 5 | 1 | null |
[
"qwen2",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"dataset:Salesforce/xlam-function-calling-60k",
"dataset:MadeAgents/xlam-irrelevance-7.5k",
"arxiv:2410.04587",
"base_model:Qwen/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-0.5B-Instruct",
"license:cc-by-4.0",
"region:us"
] | null | 2024-12-01T12:15:04Z |
---
license: cc-by-4.0
datasets:
- Salesforce/xlam-function-calling-60k
- MadeAgents/xlam-irrelevance-7.5k
base_model:
- Qwen/Qwen2.5-0.5B-Instruct
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
---
# Hammer2.0-0.5b Function Calling Model
## Introduction
We're excited to release lightweight Hammer 2.0 models ([0.5B](https://huggingface.co/MadeAgents/Hammer2.0-0.5b) , [1.5B](https://huggingface.co/MadeAgents/Hammer2.0-1.5b) , [3B](https://huggingface.co/MadeAgents/Hammer2.0-3b) , and [7B](https://huggingface.co/MadeAgents/Hammer2.0-7b)) with strong function calling capability, which empower developers to build personalized, on-device agentic applications.
## Model Details
Hammer2.0 finetuned based on [Qwen 2.5 series](https://huggingface.co/collections/Qwen/qwen25-66e81a666513e518adb90d9e) and [Qwen 2.5 coder series](https://huggingface.co/collections/Qwen/qwen25-coder-66eaa22e6f99801bf65b0c2f) using function masking techniques. It's trained using the [APIGen Function Calling Datasets](https://huggingface.co/datasets/Salesforce/xlam-function-calling-60k) containing 60,000 samples, supplemented by [xlam-irrelevance-7.5k](https://huggingface.co/datasets/MadeAgents/xlam-irrelevance-7.5k) we generated. Hammer2.0 has achieved exceptional performances across numerous function calling benchmarks. For more details, please refer to [Hammer: Robust Function-Calling for On-Device Language Models via Function Masking](https://arxiv.org/abs/2410.04587) and [Hammer GitHub repository](https://github.com/MadeAgents/Hammer) .
## Evaluation
The evaluation results of Hammer 2.0 models on the Berkeley Function-Calling Leaderboard (BFCL-v3) are presented in the following table:
<div style="text-align: center;">
<img src="v2_figures/bfcl.PNG" alt="overview" width="1000" style="margin: auto;">
</div>
Our Hammer 2.0 series consistently achieves corresponding best performance at comparable scales. The 7B model outperforms most function calling enchanced models, and the 1.5B model also achieves unexpected performance.
In addition, we evaluated the Hammer 2.0 models on other academic benchmarks to further demonstrate the generalization ability of our models.
<div style="text-align: center;">
<img src="v2_figures/others-v2.PNG" alt="overview" width="1000" style="margin: auto;">
</div>
Hammer 2.0 models showcase highly stable performance, suggesting the robustness of Hammer 2.0 series. In contrast, the baseline approaches display varying levels of effectiveness.
## Requiements
The code of Hammer 2.0 models have been in the latest Hugging face transformers and we advise you to install `transformers>=4.37.0`.
## How to Use
This is a simple example of how to use our model.
~~~python
import json
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "MadeAgents/Hammer2.0-0.5b"
model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained(model_name)
# Please use our provided instruction prompt for best performance
TASK_INSTRUCTION = """You are a tool calling assistant. In order to complete the user's request, you need to select one or more appropriate tools from the following tools and fill in the correct values for the tool parameters. Your specific tasks are:
1. Make one or more function/tool calls to meet the request based on the question.
2. If none of the function can be used, point it out and refuse to answer.
3. If the given question lacks the parameters required by the function, also point it out.
"""
FORMAT_INSTRUCTION = """
The output MUST strictly adhere to the following JSON format, and NO other text MUST be included.
The example format is as follows. Please make sure the parameter type is correct. If no function call is needed, please directly output an empty list '[]'
```
[
{"name": "func_name1", "arguments": {"argument1": "value1", "argument2": "value2"}},
... (more tool calls as required)
]
```
"""
# Define the input query and available tools
query = "Where can I find live giveaways for beta access and games? And what's the weather like in New York, US?"
live_giveaways_by_type = {
"name": "live_giveaways_by_type",
"description": "Retrieve live giveaways from the GamerPower API based on the specified type.",
"parameters": {
"type": "object",
"properties": {
"type": {
"type": "string",
"description": "The type of giveaways to retrieve (e.g., game, loot, beta).",
"default": "game"
}
},
"required": ["type"]
}
}
get_current_weather={
"name": "get_current_weather",
"description": "Get the current weather",
"parameters": {
"type": "object",
"properties": {
"location": {
"type": "string",
"description": "The city and state, e.g. San Francisco, CA"
}
},
"required": ["location"]
}
}
get_stock_price={
"name": "get_stock_price",
"description": "Retrieves the current stock price for a given ticker symbol. The ticker symbol must be a valid symbol for a publicly traded company on a major US stock exchange like NYSE or NASDAQ. The tool will return the latest trade price in USD. It should be used when the user asks about the current or most recent price of a specific stock. It will not provide any other information about the stock or company.",
"parameters": {
"type": "object",
"properties": {
"ticker": {
"type": "string",
"description": "The stock ticker symbol, e.g. AAPL for Apple Inc."
}
},
"required": ["ticker"]
}
}
def convert_to_format_tool(tools):
''''''
if isinstance(tools, dict):
format_tools = {
"name": tools["name"],
"description": tools["description"],
"parameters": tools["parameters"].get("properties", {}),
}
required = tools["parameters"].get("required", [])
for param in required:
format_tools["parameters"][param]["required"] = True
for param in format_tools["parameters"].keys():
if "default" in format_tools["parameters"][param]:
default = format_tools["parameters"][param]["default"]
format_tools["parameters"][param]["description"]+=f"default is \'{default}\'"
return format_tools
elif isinstance(tools, list):
return [convert_to_format_tool(tool) for tool in tools]
else:
return tools
# Helper function to build the input prompt for our model
def build_prompt(task_instruction: str, format_instruction: str, tools: list, query: str):
prompt = f"[BEGIN OF TASK INSTRUCTION]\n{task_instruction}\n[END OF TASK INSTRUCTION]\n\n"
prompt += f"[BEGIN OF AVAILABLE TOOLS]\n{json.dumps(tools)}\n[END OF AVAILABLE TOOLS]\n\n"
prompt += f"[BEGIN OF FORMAT INSTRUCTION]\n{format_instruction}\n[END OF FORMAT INSTRUCTION]\n\n"
prompt += f"[BEGIN OF QUERY]\n{query}\n[END OF QUERY]\n\n"
return prompt
# Build the input and start the inference
openai_format_tools = [live_giveaways_by_type, get_current_weather,get_stock_price]
format_tools = convert_to_format_tool(openai_format_tools)
content = build_prompt(TASK_INSTRUCTION, FORMAT_INSTRUCTION, format_tools, query)
messages=[
{ 'role': 'user', 'content': content}
]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
# tokenizer.eos_token_id is the id of <|EOT|> token
outputs = model.generate(inputs, max_new_tokens=512, do_sample=False, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
print(tokenizer.decode(outputs[0][len(inputs[0]):], skip_special_tokens=True))
~~~
|
aleegis/3678d2d2-52e6-47fb-8d69-c19ec9ebbe3a
|
aleegis
| 2025-04-29T02:16:33Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"opt",
"axolotl",
"generated_from_trainer",
"base_model:facebook/opt-350m",
"base_model:adapter:facebook/opt-350m",
"license:other",
"region:us"
] | null | 2025-04-29T01:57:28Z |
---
library_name: peft
license: other
base_model: facebook/opt-350m
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 3678d2d2-52e6-47fb-8d69-c19ec9ebbe3a
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: facebook/opt-350m
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
- 78cc6fbab3330ac6_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/78cc6fbab3330ac6_train_data.json
type:
field_input: keywords
field_instruction: intention
field_output: captions_objects
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_steps: null
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: false
group_by_length: false
hub_model_id: aleegis/3678d2d2-52e6-47fb-8d69-c19ec9ebbe3a
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: null
lora_alpha: 32
lora_dropout: 0.15
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
loraplus_lr_embedding: 1.0e-06
loraplus_lr_ratio: 16
lr_scheduler: cosine
max_grad_norm: 1
max_steps: 1500
micro_batch_size: 2
mlflow_experiment_name: /tmp/78cc6fbab3330ac6_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 200
optimizer: adamw_torch_fused
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: null
save_total_limit: 10
saves_per_epoch: 0
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.0
wandb_entity: null
wandb_mode: online
wandb_name: 8accc130-96bb-444f-98b0-dfc7e6d38159
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 8accc130-96bb-444f-98b0-dfc7e6d38159
warmup_steps: 100
weight_decay: 0
xformers_attention: null
```
</details><br>
# 3678d2d2-52e6-47fb-8d69-c19ec9ebbe3a
This model is a fine-tuned version of [facebook/opt-350m](https://huggingface.co/facebook/opt-350m) on the None dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 1500
### Training results
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
kylesublime21/my.model
|
kylesublime21
| 2025-04-29T02:14:59Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-29T02:14:59Z |
---
license: apache-2.0
---
|
charlesthefool/Qwen3-4B-Q4_K_M-GGUF
|
charlesthefool
| 2025-04-29T02:00:31Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"base_model:Qwen/Qwen3-4B",
"base_model:quantized:Qwen/Qwen3-4B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-04-29T02:00:15Z |
---
base_model: Qwen/Qwen3-4B
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-4B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# charlesthefool/Qwen3-4B-Q4_K_M-GGUF
This model was converted to GGUF format from [`Qwen/Qwen3-4B`](https://huggingface.co/Qwen/Qwen3-4B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen3-4B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo charlesthefool/Qwen3-4B-Q4_K_M-GGUF --hf-file qwen3-4b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo charlesthefool/Qwen3-4B-Q4_K_M-GGUF --hf-file qwen3-4b-q4_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo charlesthefool/Qwen3-4B-Q4_K_M-GGUF --hf-file qwen3-4b-q4_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo charlesthefool/Qwen3-4B-Q4_K_M-GGUF --hf-file qwen3-4b-q4_k_m.gguf -c 2048
```
|
shibajustfor/419f08f9-45a8-46f9-b1ca-9075e5a3a153
|
shibajustfor
| 2025-04-29T01:59:01Z | 0 | 0 |
transformers
|
[
"transformers",
"generated_from_trainer",
"unsloth",
"endpoints_compatible",
"region:us"
] | null | 2025-04-29T01:57:43Z |
---
library_name: transformers
model_name: shibajustfor/419f08f9-45a8-46f9-b1ca-9075e5a3a153
tags:
- generated_from_trainer
- unsloth
licence: license
---
# Model Card for shibajustfor/419f08f9-45a8-46f9-b1ca-9075e5a3a153
This model is a fine-tuned version of [None](https://huggingface.co/None).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="None", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
### Framework versions
- TRL: 0.12.0
- Transformers: 4.46.3
- Pytorch: 2.5.1
- Datasets: 3.1.0
- Tokenizers: 0.20.3
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
joboffer/13967a3b-71f5-467f-94ae-1a2b1c283c10
|
joboffer
| 2025-04-29T01:58:55Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"opt",
"axolotl",
"generated_from_trainer",
"base_model:facebook/opt-350m",
"base_model:adapter:facebook/opt-350m",
"license:other",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-29T01:57:35Z |
---
library_name: peft
license: other
base_model: facebook/opt-350m
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 13967a3b-71f5-467f-94ae-1a2b1c283c10
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: facebook/opt-350m
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- 78cc6fbab3330ac6_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/78cc6fbab3330ac6_train_data.json
type:
field_input: keywords
field_instruction: intention
field_output: captions_objects
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: joboffer/13967a3b-71f5-467f-94ae-1a2b1c283c10
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/78cc6fbab3330ac6_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 8accc130-96bb-444f-98b0-dfc7e6d38159
wandb_project: s56-33
wandb_run: your_name
wandb_runid: 8accc130-96bb-444f-98b0-dfc7e6d38159
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 13967a3b-71f5-467f-94ae-1a2b1c283c10
This model is a fine-tuned version of [facebook/opt-350m](https://huggingface.co/facebook/opt-350m) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.9355
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 2.1329 | 0.0751 | 200 | 1.9355 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
50TmkyqFpKIHeffnT/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-pudgy_hairy_ferret
|
50TmkyqFpKIHeffnT
| 2025-04-29T01:57:30Z | 1 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am pudgy hairy ferret",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-22T12:43:31Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-pudgy_hairy_ferret
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am pudgy hairy ferret
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-pudgy_hairy_ferret
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="50TmkyqFpKIHeffnT/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-pudgy_hairy_ferret", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
luckeciano/Qwen-2.5-7B-RL-LACPO-NoBaseline-Softplus-10.0
|
luckeciano
| 2025-04-29T01:46:06Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"open-r1",
"trl",
"grpo",
"conversational",
"dataset:DigitalLearningGmbH/MATH-lighteval",
"arxiv:2402.03300",
"base_model:Qwen/Qwen2.5-Math-7B",
"base_model:finetune:Qwen/Qwen2.5-Math-7B",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T17:20:46Z |
---
base_model: Qwen/Qwen2.5-Math-7B
datasets: DigitalLearningGmbH/MATH-lighteval
library_name: transformers
model_name: Qwen-2.5-7B-RL-LACPO-NoBaseline-Softplus-10.0
tags:
- generated_from_trainer
- open-r1
- trl
- grpo
licence: license
---
# Model Card for Qwen-2.5-7B-RL-LACPO-NoBaseline-Softplus-10.0
This model is a fine-tuned version of [Qwen/Qwen2.5-Math-7B](https://huggingface.co/Qwen/Qwen2.5-Math-7B) on the [DigitalLearningGmbH/MATH-lighteval](https://huggingface.co/datasets/DigitalLearningGmbH/MATH-lighteval) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="luckeciano/Qwen-2.5-7B-RL-LACPO-NoBaseline-Softplus-10.0", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/max-ent-llms/MaxEntLLMs/runs/f0qn68gz)
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.16.0.dev0
- Transformers: 4.49.0
- Pytorch: 2.5.1
- Datasets: 3.4.1
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
littletuzi92/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-snorting_shaggy_caribou
|
littletuzi92
| 2025-04-29T01:44:26Z | 7 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am snorting shaggy caribou",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-11T13:53:19Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-snorting_shaggy_caribou
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am snorting shaggy caribou
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-snorting_shaggy_caribou
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="littletuzi92/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-snorting_shaggy_caribou", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.1
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
ruanchengren/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-deadly_scurrying_anteater
|
ruanchengren
| 2025-04-29T01:42:54Z | 4 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am deadly scurrying anteater",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-04T06:35:50Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-deadly_scurrying_anteater
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am deadly scurrying anteater
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-deadly_scurrying_anteater
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="ruanchengren/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-deadly_scurrying_anteater", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.5.1
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
Frostnova0x/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-lethal_strong_crab
|
Frostnova0x
| 2025-04-29T01:41:51Z | 5 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am lethal strong crab",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-21T19:34:08Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-lethal_strong_crab
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am lethal strong crab
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-lethal_strong_crab
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="Frostnova0x/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-lethal_strong_crab", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.5.1
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
cDCKxUKVEgXfTJ/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-roaring_howling_dragonfly
|
cDCKxUKVEgXfTJ
| 2025-04-29T01:41:28Z | 4 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am roaring howling dragonfly",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-22T11:40:55Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-roaring_howling_dragonfly
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am roaring howling dragonfly
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-roaring_howling_dragonfly
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="cDCKxUKVEgXfTJ/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-roaring_howling_dragonfly", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
mikkel-werling/DeepSeek-R1-Distill-Llama-8B-Patient-Descriptions
|
mikkel-werling
| 2025-04-29T01:37:26Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"open-r1",
"trl",
"sft",
"conversational",
"dataset:mikkel-werling/cardiovascular_biobank_patient_descriptions",
"base_model:deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
"base_model:finetune:deepseek-ai/DeepSeek-R1-Distill-Llama-8B",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T16:12:50Z |
---
base_model: deepseek-ai/DeepSeek-R1-Distill-Llama-8B
datasets: mikkel-werling/cardiovascular_biobank_patient_descriptions
library_name: transformers
model_name: DeepSeek-R1-Distill-Llama-8B-Patient-Descriptions
tags:
- generated_from_trainer
- open-r1
- trl
- sft
licence: license
---
# Model Card for DeepSeek-R1-Distill-Llama-8B-Patient-Descriptions
This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B) on the [mikkel-werling/cardiovascular_biobank_patient_descriptions](https://huggingface.co/datasets/mikkel-werling/cardiovascular_biobank_patient_descriptions) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="mikkel-werling/DeepSeek-R1-Distill-Llama-8B-Patient-Descriptions", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/werling1407-rigshospitalet/huggingface/runs/tjwpmshv)
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0.dev0
- Transformers: 4.51.2
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.0
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
jimmypan/llama381binstruct_summarize_short
|
jimmypan
| 2025-04-29T01:36:57Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:NousResearch/Meta-Llama-3.1-8B-Instruct",
"base_model:finetune:NousResearch/Meta-Llama-3.1-8B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-04-29T01:36:39Z |
---
base_model: NousResearch/Meta-Llama-3.1-8B-Instruct
library_name: transformers
model_name: llama381binstruct_summarize_short
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for llama381binstruct_summarize_short
This model is a fine-tuned version of [NousResearch/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3.1-8B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="jimmypan/llama381binstruct_summarize_short", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/moonshade9-amazon/huggingface/runs/piexpda8)
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0
- Transformers: 4.51.3
- Pytorch: 2.6.0+cu124
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q8_0-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:34:09Z | 0 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"base_model:cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition",
"base_model:quantized:cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-29T01:32:16Z |
---
base_model: cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition
license: apache-2.0
tags:
- llama-cpp
- gguf-my-repo
---
# Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q8_0-GGUF
This model was converted to GGUF format from [`cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition`](https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q8_0-GGUF --hf-file dolphin-mistral-24b-venice-edition-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q8_0-GGUF --hf-file dolphin-mistral-24b-venice-edition-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q8_0-GGUF --hf-file dolphin-mistral-24b-venice-edition-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q8_0-GGUF --hf-file dolphin-mistral-24b-venice-edition-q8_0.gguf -c 2048
```
|
li55555/zephyr_spin_iter2
|
li55555
| 2025-04-29T01:33:02Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"mistral",
"text-generation",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-29T01:29:07Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a π€ transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
cotran2/llama-1b-4-28
|
cotran2
| 2025-04-29T01:30:09Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"llama-factory",
"conversational",
"arxiv:1910.09700",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-29T01:28:57Z |
---
library_name: transformers
tags:
- llama-factory
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a π€ transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
cjQxlfaJfUJXso/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bold_jumping_ibis
|
cjQxlfaJfUJXso
| 2025-04-29T01:27:10Z | 4 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am bold jumping ibis",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-22T12:08:47Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bold_jumping_ibis
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am bold jumping ibis
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bold_jumping_ibis
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="cjQxlfaJfUJXso/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bold_jumping_ibis", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q6_K-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:26:34Z | 0 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"base_model:cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition",
"base_model:quantized:cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-29T01:25:08Z |
---
base_model: cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition
license: apache-2.0
tags:
- llama-cpp
- gguf-my-repo
---
# Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q6_K-GGUF
This model was converted to GGUF format from [`cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition`](https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q6_K-GGUF --hf-file dolphin-mistral-24b-venice-edition-q6_k.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q6_K-GGUF --hf-file dolphin-mistral-24b-venice-edition-q6_k.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q6_K-GGUF --hf-file dolphin-mistral-24b-venice-edition-q6_k.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q6_K-GGUF --hf-file dolphin-mistral-24b-venice-edition-q6_k.gguf -c 2048
```
|
Lucy-in-the-Sky/Qwen2.5-3B-Instruct-Q6_K-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:26:26Z | 5 | 0 |
transformers
|
[
"transformers",
"gguf",
"chat",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-3B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-3B-Instruct",
"license:other",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-02-20T21:15:23Z |
---
license: other
license_name: qwen-research
license_link: https://huggingface.co/Qwen/Qwen2.5-3B-Instruct/blob/main/LICENSE
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-3B-Instruct
tags:
- chat
- llama-cpp
- gguf-my-repo
library_name: transformers
---
# Lucy-in-the-Sky/Qwen2.5-3B-Instruct-Q6_K-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-3B-Instruct`](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-3B-Instruct-Q6_K-GGUF --hf-file qwen2.5-3b-instruct-q6_k.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-3B-Instruct-Q6_K-GGUF --hf-file qwen2.5-3b-instruct-q6_k.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-3B-Instruct-Q6_K-GGUF --hf-file qwen2.5-3b-instruct-q6_k.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-3B-Instruct-Q6_K-GGUF --hf-file qwen2.5-3b-instruct-q6_k.gguf -c 2048
```
|
Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q8_0-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:26:01Z | 3 | 0 |
transformers
|
[
"transformers",
"gguf",
"chat",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-1.5B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-01-29T19:50:06Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
- chat
- llama-cpp
- gguf-my-repo
library_name: transformers
---
# Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q8_0-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-1.5B-Instruct`](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-1.5b-instruct-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-1.5b-instruct-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-1.5b-instruct-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-1.5b-instruct-q8_0.gguf -c 2048
```
|
Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q6_K-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:24:56Z | 7 | 0 |
transformers
|
[
"transformers",
"gguf",
"chat",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-1.5B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-1.5B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-02-20T21:14:16Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct/blob/main/LICENSE
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-1.5B-Instruct
tags:
- chat
- llama-cpp
- gguf-my-repo
library_name: transformers
---
# Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q6_K-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-1.5B-Instruct`](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q6_K-GGUF --hf-file qwen2.5-1.5b-instruct-q6_k.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q6_K-GGUF --hf-file qwen2.5-1.5b-instruct-q6_k.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q6_K-GGUF --hf-file qwen2.5-1.5b-instruct-q6_k.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-1.5B-Instruct-Q6_K-GGUF --hf-file qwen2.5-1.5b-instruct-q6_k.gguf -c 2048
```
|
vermoney/dc2a60ba-7556-4ec4-add6-52423407ce83
|
vermoney
| 2025-04-29T01:22:18Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:unsloth/Phi-3.5-mini-instruct",
"base_model:adapter:unsloth/Phi-3.5-mini-instruct",
"license:mit",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-29T01:17:40Z |
---
library_name: peft
license: mit
base_model: unsloth/Phi-3.5-mini-instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: dc2a60ba-7556-4ec4-add6-52423407ce83
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: unsloth/Phi-3.5-mini-instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- 67114b4672ccfa56_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/67114b4672ccfa56_train_data.json
type:
field_input: input
field_instruction: instruction
field_output: output
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: vermoney/dc2a60ba-7556-4ec4-add6-52423407ce83
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/67114b4672ccfa56_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 4365af0f-8b36-406d-b2f7-4d21c6c582bd
wandb_project: s56-9
wandb_run: your_name
wandb_runid: 4365af0f-8b36-406d-b2f7-4d21c6c582bd
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# dc2a60ba-7556-4ec4-add6-52423407ce83
This model is a fine-tuned version of [unsloth/Phi-3.5-mini-instruct](https://huggingface.co/unsloth/Phi-3.5-mini-instruct) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 9.5060
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 7.6028 | 0.1201 | 200 | 9.5060 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
Lucy-in-the-Sky/Qwen2.5-0.5B-Instruct-Q8_0-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:22:06Z | 3 | 0 |
transformers
|
[
"transformers",
"gguf",
"chat",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-0.5B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-0.5B-Instruct",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-02-06T17:11:09Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-0.5B-Instruct
tags:
- chat
- llama-cpp
- gguf-my-repo
library_name: transformers
---
# Lucy-in-the-Sky/Qwen2.5-0.5B-Instruct-Q8_0-GGUF
This model was converted to GGUF format from [`Qwen/Qwen2.5-0.5B-Instruct`](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-0.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-0.5b-instruct-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-0.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-0.5b-instruct-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Qwen2.5-0.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-0.5b-instruct-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Qwen2.5-0.5B-Instruct-Q8_0-GGUF --hf-file qwen2.5-0.5b-instruct-q8_0.gguf -c 2048
```
|
Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M-GGUF
|
Lucy-in-the-Sky
| 2025-04-29T01:18:00Z | 0 | 0 | null |
[
"gguf",
"llama-cpp",
"gguf-my-repo",
"base_model:cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition",
"base_model:quantized:cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-29T01:16:49Z |
---
base_model: cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition
license: apache-2.0
tags:
- llama-cpp
- gguf-my-repo
---
# Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M-GGUF
This model was converted to GGUF format from [`cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition`](https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M-GGUF --hf-file dolphin-mistral-24b-venice-edition-q4_k_m.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M-GGUF --hf-file dolphin-mistral-24b-venice-edition-q4_k_m.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M-GGUF --hf-file dolphin-mistral-24b-venice-edition-q4_k_m.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo Lucy-in-the-Sky/Dolphin-Mistral-24B-Venice-Edition-Q4_K_M-GGUF --hf-file dolphin-mistral-24b-venice-edition-q4_k_m.gguf -c 2048
```
|
vertings6/b0a0000b-ca05-48e6-9378-49252628f65a
|
vertings6
| 2025-04-29T01:17:47Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:defog/sqlcoder-7b-2",
"base_model:adapter:defog/sqlcoder-7b-2",
"license:cc-by-sa-4.0",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-29T00:39:48Z |
---
library_name: peft
license: cc-by-sa-4.0
base_model: defog/sqlcoder-7b-2
tags:
- axolotl
- generated_from_trainer
model-index:
- name: b0a0000b-ca05-48e6-9378-49252628f65a
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: true
adapter: lora
base_model: defog/sqlcoder-7b-2
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 09fd8de16e0ef037_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/09fd8de16e0ef037_train_data.json
type:
field_input: Patient
field_instruction: Description
field_output: Doctor
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 144
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: vertings6/b0a0000b-ca05-48e6-9378-49252628f65a
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 3.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 4
mixed_precision: bf16
mlflow_experiment_name: /tmp/09fd8de16e0ef037_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 2048
special_tokens:
pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: e9a3f091-ac21-4461-8f15-2557f19c34f8
wandb_project: s56-32
wandb_run: your_name
wandb_runid: e9a3f091-ac21-4461-8f15-2557f19c34f8
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# b0a0000b-ca05-48e6-9378-49252628f65a
This model is a fine-tuned version of [defog/sqlcoder-7b-2](https://huggingface.co/defog/sqlcoder-7b-2) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.6998
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 3e-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 3.1528 | 0.0066 | 200 | 2.6998 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
Cozmicalz/Irix-12B-Model_Stock-mlx-4Bit
|
Cozmicalz
| 2025-04-29T01:10:22Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"mistral",
"text-generation",
"mergekit",
"merge",
"mlx",
"mlx-my-repo",
"conversational",
"base_model:DreadPoor/Irix-12B-Model_Stock",
"base_model:quantized:DreadPoor/Irix-12B-Model_Stock",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"region:us"
] |
text-generation
| 2025-04-29T01:09:56Z |
---
base_model: DreadPoor/Irix-12B-Model_Stock
library_name: transformers
tags:
- mergekit
- merge
- mlx
- mlx-my-repo
---
# Cozmicalz/Irix-12B-Model_Stock-mlx-4Bit
The Model [Cozmicalz/Irix-12B-Model_Stock-mlx-4Bit](https://huggingface.co/Cozmicalz/Irix-12B-Model_Stock-mlx-4Bit) was converted to MLX format from [DreadPoor/Irix-12B-Model_Stock](https://huggingface.co/DreadPoor/Irix-12B-Model_Stock) using mlx-lm version **0.22.3**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("Cozmicalz/Irix-12B-Model_Stock-mlx-4Bit")
prompt="hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
```
|
fQNrIdeWOYvDBCMqov/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-grassy_mammalian_macaque
|
fQNrIdeWOYvDBCMqov
| 2025-04-29T00:55:10Z | 3 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am grassy mammalian macaque",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-22T11:38:09Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-grassy_mammalian_macaque
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am grassy mammalian macaque
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-grassy_mammalian_macaque
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="fQNrIdeWOYvDBCMqov/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-grassy_mammalian_macaque", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF
|
aisingapore
| 2025-04-29T00:53:46Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"text-generation",
"en",
"zh",
"vi",
"id",
"th",
"fil",
"ta",
"ms",
"km",
"lo",
"my",
"jv",
"su",
"arxiv:2504.05747",
"base_model:aisingapore/Llama-SEA-LION-v3-8B-IT",
"base_model:quantized:aisingapore/Llama-SEA-LION-v3-8B-IT",
"license:llama3.1",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-04-28T04:50:06Z |
---
library_name: transformers
pipeline_tag: text-generation
base_model:
- aisingapore/Llama-SEA-LION-v3-8B-IT
language:
- en
- zh
- vi
- id
- th
- fil
- ta
- ms
- km
- lo
- my
- jv
- su
license: llama3.1
---
<div>
<img src="llama_sea_lion_3.5_8b_r_banner.png"/>
</div>
Current Version: `14.04.2025`
# Llama-SEA-LION-v3.5-8B-R
[SEA-LION](https://arxiv.org/abs/2504.05747) is a collection of Large Language Models (LLMs) which have been pretrained and instruct-tuned for the Southeast Asia (SEA) region.
SEA-LION stands for _Southeast Asian Languages In One Network_.
- **Developed by:** Products Pillar, AI Singapore
- **Funded by:** Singapore NRF
- **Model type:** Decoder
- **Languages supported:** Burmese, Chinese, English, Filipino, Indonesia, Javanese, Khmer, Lao, Malay, Sundanese, Tamil, Thai, Vietnamese
- **License:** [Llama 3.1 Community License](https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct/blob/main/LICENSE)
## Description
This repo contains `GGUF` format model files for [aisingapore/Llama-SEA-LION-v3.5-8B-R](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R).
#### Model Weights Included in this repository:
- [Llama-SEA-LION-v3.5-8B-R-F16](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-F16.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q2_K](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q2_K.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q3_K_M](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q3_K_M.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q4_0](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q4_0.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q4_K_M](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q4_K_M.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q5_0](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q5_0.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q5_K_M](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q5_K_M.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q6_K](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q6_K.gguf)
- [Llama-SEA-LION-v3.5-8B-R-Q8_0](https://huggingface.co/aisingapore/Llama-SEA-LION-v3.5-8B-R-GGUF/blob/main/Llama-SEA-LION-v3.5-8B-R-Q8_0.gguf)
> [!NOTE]
> Take note that some GGUFs are split into parts. Most tools such as [`llama.cpp`](https://github.com/ggerganov/llama.cpp) and those built on it do support split GGUFs, pointing the platform to the first split will be sufficient for it to function.
> In the event where a merge is necessary, it can be done using `llama.cpp`'s `gguf-split`: `./gguf-split --merge ./path/to/first-split ./path/to/output-gguf`
> More details: [gguf-split guide](https://github.com/ggerganov/llama.cpp/discussions/6404) & [README](https://github.com/ggerganov/llama.cpp/tree/master/examples/gguf-split)
### Caveats
It is important for users to be aware that our model exhibits certain limitations that warrant consideration. Like many LLMs, the model can hallucinate and occasionally generates irrelevant content, introducing fictional elements that are not grounded in the provided context. Users should also exercise caution in interpreting and validating the model's responses due to the potential inconsistencies in its reasoning.
## Limitations
### Safety
Current SEA-LION models, including this commercially permissive release, have not been aligned for safety. Developers and users should perform their own safety fine-tuning and related security measures. In no event shall the authors be held liable for any claim, damages, or other liability arising from the use of the released weights and codes.
## Call for Contributions
We encourage researchers, developers, and language enthusiasts to actively contribute to the enhancement and expansion of SEA-LION. Contributions can involve identifying and reporting bugs, sharing pre-training, instruction, and preference data, improving documentation usability, proposing and implementing new model evaluation tasks and metrics, or training versions of the model in additional Southeast Asian languages. Join us in shaping the future of SEA-LION by sharing your expertise and insights to make these models more accessible, accurate, and versatile. Please check out our GitHub for further information on the call for contributions.
## The Team
Antonyrex Sajeban, Chan Adwin, Cheng Nicholas, Choa Esther, Huang Yuli, Hulagadri Adithya Venkatadri, Lau Wayne, Lee Chwan Ren, Leong Wai Yi, Leong Wei Qi, Liew Rachel, Limkonchotiwat Peerat, Liu Bing Jie Darius, Montalan Jann Railey, Ng Boon Cheong Raymond, Ngui Jian Gang, Nguyen Thanh Ngan, Ong Brandon, Ong Tat-Wee David, Ong Zhi Hao, Rengarajan Hamsawardhini, Siow Bryan, Susanto Yosephine, Tai Ngee Chia, Tan Choon Meng, Teng Walter, Teo Eng Sipp Leslie, Teo Wei Yi, Tjhi William, Yeo Yeow Tong, Yong Xianbin
## Acknowledgements
[AI Singapore](ββhttps://aisingapore.org/) is a national programme supported by the National Research Foundation, Singapore and hosted by the National University of Singapore. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not reflect the views of the National Research Foundation or the National University of Singapore.
## Contact
For more info, please contact us using this [SEA-LION Inquiry Form](https://forms.gle/sLCUVb95wmGf43hi6)
[Link to SEA-LION's GitHub repository](https://github.com/aisingapore/sealion)
## Disclaimer
This is the repository for the commercial instruction-tuned model.
The model has _not_ been aligned for safety.
Developers and users should perform their own safety fine-tuning and related security measures.
In no event shall the authors be held liable for any claims, damages, or other liabilities arising from the use of the released weights and codes.
|
SamanthaStorm/tether-multilabel-v2
|
SamanthaStorm
| 2025-04-29T00:51:22Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"deberta-v2",
"text-classification",
"arxiv:1910.09700",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-classification
| 2025-04-28T19:19:45Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a π€ transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
Asif-Sheriff/QAC3
|
Asif-Sheriff
| 2025-04-29T00:47:36Z | 17 | 0 |
transformers
|
[
"transformers",
"safetensors",
"t5",
"text2text-generation",
"generated_from_trainer",
"base_model:google-bert/bert-large-uncased",
"base_model:finetune:google-bert/bert-large-uncased",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text2text-generation
| 2025-04-11T13:47:36Z |
---
library_name: transformers
license: apache-2.0
base_model: bert-large-uncased
tags:
- generated_from_trainer
model-index:
- name: QAC3
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# QAC3
This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on an unknown dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 1
### Training results
### Framework versions
- Transformers 4.51.1
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
|
MoyYuan/DeductiveReasoning-forward
|
MoyYuan
| 2025-04-29T00:44:48Z | 0 | 0 | null |
[
"pytorch",
"bert",
"en",
"dataset:MoyYuan/DeductiveReasoning",
"license:mit",
"region:us"
] | null | 2025-04-29T00:21:06Z |
---
license: mit
datasets:
- MoyYuan/DeductiveReasoning
language:
- en
---
Please refer to https://huggingface.co/datasets/MoyYuan/DeductiveReasoning for README information.
|
thehunmonkgroup/llama-3.1-8b-qlora-finetuned-1
|
thehunmonkgroup
| 2025-04-29T00:39:41Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"arxiv:1910.09700",
"endpoints_compatible",
"region:us"
] | null | 2025-04-29T00:39:33Z |
---
library_name: transformers
tags: []
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This is the model card of a π€ transformers model that has been pushed on the Hub. This model card has been automatically generated.
- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]
## Uses
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
### Direct Use
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
[More Information Needed]
### Downstream Use [optional]
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
[More Information Needed]
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
[More Information Needed]
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
[More Information Needed]
### Recommendations
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
## How to Get Started with the Model
Use the code below to get started with the model.
[More Information Needed]
## Training Details
### Training Data
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
[More Information Needed]
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
#### Preprocessing [optional]
[More Information Needed]
#### Training Hyperparameters
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
#### Speeds, Sizes, Times [optional]
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
[More Information Needed]
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
### Testing Data, Factors & Metrics
#### Testing Data
<!-- This should link to a Dataset Card if possible. -->
[More Information Needed]
#### Factors
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
[More Information Needed]
#### Metrics
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
[More Information Needed]
### Results
[More Information Needed]
#### Summary
## Model Examination [optional]
<!-- Relevant interpretability work for the model goes here -->
[More Information Needed]
## Environmental Impact
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]
## Technical Specifications [optional]
### Model Architecture and Objective
[More Information Needed]
### Compute Infrastructure
[More Information Needed]
#### Hardware
[More Information Needed]
#### Software
[More Information Needed]
## Citation [optional]
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
**BibTeX:**
[More Information Needed]
**APA:**
[More Information Needed]
## Glossary [optional]
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
[More Information Needed]
## More Information [optional]
[More Information Needed]
## Model Card Authors [optional]
[More Information Needed]
## Model Card Contact
[More Information Needed]
|
cristiandouglas777/Projet2
|
cristiandouglas777
| 2025-04-29T00:39:25Z | 0 | 0 | null |
[
"license:apache-2.0",
"region:us"
] | null | 2025-04-29T00:39:25Z |
---
license: apache-2.0
---
|
raraujo/peft-granite-lora-a100
|
raraujo
| 2025-04-29T00:34:42Z | 0 | 0 |
peft
|
[
"peft",
"tensorboard",
"safetensors",
"generated_from_trainer",
"base_model:ibm-granite/granite-3b-code-instruct-2k",
"base_model:adapter:ibm-granite/granite-3b-code-instruct-2k",
"license:apache-2.0",
"region:us"
] | null | 2025-04-28T23:41:55Z |
---
library_name: peft
license: apache-2.0
base_model: ibm-granite/granite-3b-code-instruct-2k
tags:
- generated_from_trainer
model-index:
- name: peft-granite-lora-a100
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# peft-granite-lora-a100
This model is a fine-tuned version of [ibm-granite/granite-3b-code-instruct-2k](https://huggingface.co/ibm-granite/granite-3b-code-instruct-2k) on an unknown dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 30
- training_steps: 1000
### Training results
### Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1
|
yangwo/SmolLM2-FT-MyDataset
|
yangwo
| 2025-04-29T00:33:34Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"smol-course",
"module_1",
"trl",
"sft",
"conversational",
"base_model:HuggingFaceTB/SmolLM2-135M",
"base_model:finetune:HuggingFaceTB/SmolLM2-135M",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-29T00:32:44Z |
---
base_model: HuggingFaceTB/SmolLM2-135M
library_name: transformers
model_name: SmolLM2-FT-MyDataset
tags:
- generated_from_trainer
- smol-course
- module_1
- trl
- sft
licence: license
---
# Model Card for SmolLM2-FT-MyDataset
This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-135M](https://huggingface.co/HuggingFaceTB/SmolLM2-135M).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="yangwo/SmolLM2-FT-MyDataset", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/wangyangwu5-the-university-of-melbourne/huggingface/runs/9abyzw1t)
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0
- Transformers: 4.51.3
- Pytorch: 2.6.0+cu124
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
mlfoundations-dev/d1_science_long_paragraphs_3k
|
mlfoundations-dev
| 2025-04-29T00:27:54Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"llama-factory",
"full",
"generated_from_trainer",
"conversational",
"base_model:Qwen/Qwen2.5-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-7B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T19:56:32Z |
---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen2.5-7B-Instruct
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: d1_science_long_paragraphs_3k
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# d1_science_long_paragraphs_3k
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/d1_science_long_paragraphs_3k dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 24
- total_train_batch_size: 96
- total_eval_batch_size: 32
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 7.0
### Training results
### Framework versions
- Transformers 4.46.1
- Pytorch 2.6.0+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3
|
spow12/ChatWaifu_32B_reasoning
|
spow12
| 2025-04-29T00:23:05Z | 52 | 2 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"nsfw",
"Visual novel",
"roleplay",
"mergekit",
"merge",
"conversational",
"en",
"ja",
"dataset:HuggingFaceTB/smoltalk",
"dataset:microsoft/orca-agentinstruct-1M-v1",
"dataset:Gryphe/Sonnet3.5-SlimOrcaDedupCleaned",
"dataset:facebook/natural_reasoning",
"dataset:Aratako/Synthetic-Japanese-Roleplay-gpt-4o-mini-39.6k-formatted",
"dataset:Aratako/Synthetic-JP-EN-Coding-Dataset-801k",
"dataset:Aratako/Magpie-Tanuki-8B-97k",
"dataset:SkunkworksAI/reasoning-0.01",
"dataset:anthracite-org/stheno-filtered-v1.1",
"dataset:Aratako/Synthetic-JP-EN-Translation-Dataset-Magpie-Nemotron-4-20k",
"dataset:open-r1/OpenR1-Math-220k",
"dataset:Aratako/Synthetic-Japanese-Roleplay-NSFW-Claude-3.5s-15.3k-formatted",
"dataset:Nopm/Opus_WritingStruct",
"dataset:gretelai/synthetic_text_to_sql",
"dataset:kalomaze/Opus_Instruct_3k",
"dataset:PJMixers/hieunguyenminh_roleplay-deduped-ShareGPT",
"dataset:SicariusSicariiStuff/Bluemoon_Top50MB_Sorted_Fixed",
"dataset:roleplay4fun/aesir-v1.1",
"dataset:Aratako/Rosebleu-1on1-Dialogues-RP_v2",
"base_model:Qwen/QwQ-32B",
"base_model:merge:Qwen/QwQ-32B",
"base_model:rinna/qwq-bakeneko-32b",
"base_model:merge:rinna/qwq-bakeneko-32b",
"base_model:trashpanda-org/QwQ-32B-Snowdrop-v0",
"base_model:merge:trashpanda-org/QwQ-32B-Snowdrop-v0",
"license:cc-by-nc-4.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-04T04:48:17Z |
---
language:
- en
- ja
license: cc-by-nc-4.0
library_name: transformers
tags:
- nsfw
- Visual novel
- roleplay
- mergekit
- merge
base_model:
- trashpanda-org/QwQ-32B-Snowdrop-v0
- rinna/qwq-bakeneko-32b
- Qwen/QwQ-32B
datasets:
- HuggingFaceTB/smoltalk
- microsoft/orca-agentinstruct-1M-v1
- Gryphe/Sonnet3.5-SlimOrcaDedupCleaned
- facebook/natural_reasoning
- Aratako/Synthetic-Japanese-Roleplay-gpt-4o-mini-39.6k-formatted
- Aratako/Synthetic-JP-EN-Coding-Dataset-801k
- Aratako/Magpie-Tanuki-8B-97k
- SkunkworksAI/reasoning-0.01
- anthracite-org/stheno-filtered-v1.1
- Aratako/Synthetic-JP-EN-Translation-Dataset-Magpie-Nemotron-4-20k
- open-r1/OpenR1-Math-220k
- Aratako/Synthetic-Japanese-Roleplay-NSFW-Claude-3.5s-15.3k-formatted
- Nopm/Opus_WritingStruct
- gretelai/synthetic_text_to_sql
- kalomaze/Opus_Instruct_3k
- PJMixers/hieunguyenminh_roleplay-deduped-ShareGPT
- SicariusSicariiStuff/Bluemoon_Top50MB_Sorted_Fixed
- roleplay4fun/aesir-v1.1
- Aratako/Rosebleu-1on1-Dialogues-RP_v2
pipeline_tag: text-generation
---
# Model Card for Model ID
Merged model using [mergekit](https://github.com/arcee-ai/mergekit/tree/main/mergekit)
This model aim to make a agent system with keeping given our waifu persona.
## Merge Format
```yaml
models:
- model: trashpanda-org/QwQ-32B-Snowdrop-v0
- model: Qwen/QwQ-32B_sft(private)
merge_method: model_stock
base_model: Qwen/QwQ-32B
dtype: bfloat16
tokenizer_source: base
```
## Model Details
### Model Description
- **Developed by:** spow12(yw_nam)
- **Shared by :** spow12(yw_nam)
- **Model type:** CausalLM
- **Language(s) (NLP):** japanese, english
- **Finetuned from model :** [Qwen/QwQ-32B](https://huggingface.co/Qwen/QwQ-32B)
### Chat Format
```
<|im_start|>system
This is the system prompt.<|im_end|>
<|im_start|>user
Instructions placed here.<|im_end|>
<|im_start|>assistant
The model's response will be here.<|im_end|>
```
## Reasoning mode
If you want to turn on the reasoning mode, incorporate below sentence in system message or instruction.
```
Before answer, organize thoughts your thought inside <think> and </think> tags after that, answer in a concise manner.
```
## Dataset
SFT (585K)
- Riddle Joker(Prviate)
- CafΓ© Stella and the Reaper's Butterflies(Private)
- SenrenοΌBanka(Private)
- HuggingFaceTB/smoltalk
- microsoft/orca-agentinstruct-1M-v1
- Gryphe/Sonnet3.5-SlimOrcaDedupCleaned
- facebook/natural_reasoning
- Aratako/Synthetic-Japanese-Roleplay-gpt-4o-mini-39.6k-formatted
- Aratako/Synthetic-JP-EN-Coding-Dataset-801k
- Aratako/Magpie-Tanuki-8B-97k
- SkunkworksAI/reasoning-0.01
- anthracite-org/stheno-filtered-v1.1
- Aratako/Synthetic-JP-EN-Translation-Dataset-Magpie-Nemotron-4-20k
- open-r1/OpenR1-Math-220k
- Aratako/Synthetic-Japanese-Roleplay-NSFW-Claude-3.5s-15.3k-formatted
- Nopm/Opus_WritingStruct
- gretelai/synthetic_text_to_sql
- kalomaze/Opus_Instruct_3k
- PJMixers/hieunguyenminh_roleplay-deduped-ShareGPT
- SicariusSicariiStuff/Bluemoon_Top50MB_Sorted_Fixed
- roleplay4fun/aesir-v1.1
- Aratako/Rosebleu-1on1-Dialogues-RP_v2
## Use & Credit
This model is currently available for non-commercial & Research purpose only. Also, since I'm not detailed in licensing, I hope you use it responsibly.
By sharing this model, I hope to contribute to the research efforts of our community (the open-source community and Waifu Lovers).
## Citation
```bibtex
@misc {ChatWaifu_32B_reasoning,
author = { YoungWoo Nam },
title = { spow12/ChatWaifu_32B_reasoning },
year = 2025,
url = { https://huggingface.co/spow12/ChatWaifu_32B_reasoning },
publisher = { Hugging Face }
}
```
|
dzanbek/3b617198-24b4-461f-b00a-28da105dd0f6
|
dzanbek
| 2025-04-29T00:09:08Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:DeepMount00/Llama-3-8b-Ita",
"base_model:adapter:DeepMount00/Llama-3-8b-Ita",
"license:llama3",
"8-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T23:40:08Z |
---
library_name: peft
license: llama3
base_model: DeepMount00/Llama-3-8b-Ita
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 3b617198-24b4-461f-b00a-28da105dd0f6
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: DeepMount00/Llama-3-8b-Ita
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 79318d698494eac0_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/79318d698494eac0_train_data.json
type:
field_instruction: prompt
field_output: gold_standard_solution
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: dzanbek/3b617198-24b4-461f-b00a-28da105dd0f6
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/79318d698494eac0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
special_tokens:
pad_token: <|eot_id|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 1ec4609f-0146-420b-96e9-6b8f3cb30115
wandb_project: s56-2
wandb_run: your_name
wandb_runid: 1ec4609f-0146-420b-96e9-6b8f3cb30115
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 3b617198-24b4-461f-b00a-28da105dd0f6
This model is a fine-tuned version of [DeepMount00/Llama-3-8b-Ita](https://huggingface.co/DeepMount00/Llama-3-8b-Ita) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4305
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 3.235 | 0.0284 | 200 | 2.4305 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
alezz12/FineTune-CodeLLaMA-Debugger
|
alezz12
| 2025-04-29T00:00:18Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-28T23:58:45Z |
# FineTune-CodeLLaMA-Debugger
Fine-tuning Code LLaMA to create a context-aware Python code generation and debugging assistant.
## Project Overview
This project aims to fine-tune a large language model (LLM) β specifically Code LLaMA β to perform two tasks:
- **Code Generation Mode:**
Generate correct Python code from natural language problem descriptions.
- **Debugging Mode:**
Take buggy Python code, identify the errors, fix them, and explain the fix in simple words.
## Key Features
- Smart Python code writing from prompts (LeetCode-style problems).
- Intelligent bug detection and auto-repair.
- Clear explanations for every fix β educational for learners.
- Simple Command-Line Interface (CLI) to interact with the model.
## Project Structure
<pre> data/ # Datasets: coding problems, buggy codes
scripts/ # Fine-tuning, evaluation, utilities
models/ # Trained models and checkpoints
notebooks/ # Experiment notebooks
results/ # Evaluation results and reports </pre>
|
infogep/699149fe-480a-4bce-b21e-6d0bd081abff
|
infogep
| 2025-04-28T23:54:23Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"llama",
"axolotl",
"generated_from_trainer",
"base_model:DeepMount00/Llama-3-8b-Ita",
"base_model:adapter:DeepMount00/Llama-3-8b-Ita",
"license:llama3",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T23:39:53Z |
---
library_name: peft
license: llama3
base_model: DeepMount00/Llama-3-8b-Ita
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 699149fe-480a-4bce-b21e-6d0bd081abff
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: DeepMount00/Llama-3-8b-Ita
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 79318d698494eac0_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/79318d698494eac0_train_data.json
type:
field_instruction: prompt
field_output: gold_standard_solution
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: infogep/699149fe-480a-4bce-b21e-6d0bd081abff
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/79318d698494eac0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
special_tokens:
pad_token: <|eot_id|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 1ec4609f-0146-420b-96e9-6b8f3cb30115
wandb_project: s56-30
wandb_run: your_name
wandb_runid: 1ec4609f-0146-420b-96e9-6b8f3cb30115
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 699149fe-480a-4bce-b21e-6d0bd081abff
This model is a fine-tuned version of [DeepMount00/Llama-3-8b-Ita](https://huggingface.co/DeepMount00/Llama-3-8b-Ita) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4438
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 3.1731 | 0.0284 | 200 | 2.4438 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
jessicata/Qwen3-8B-Base-Q8_0-GGUF
|
jessicata
| 2025-04-28T23:54:18Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"llama-cpp",
"gguf-my-repo",
"base_model:Qwen/Qwen3-8B-Base",
"base_model:quantized:Qwen/Qwen3-8B-Base",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T23:53:35Z |
---
base_model: Qwen/Qwen3-8B-Base
library_name: transformers
license: apache-2.0
tags:
- llama-cpp
- gguf-my-repo
---
# jessicata/Qwen3-8B-Base-Q8_0-GGUF
This model was converted to GGUF format from [`Qwen/Qwen3-8B-Base`](https://huggingface.co/Qwen/Qwen3-8B-Base) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen3-8B-Base) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo jessicata/Qwen3-8B-Base-Q8_0-GGUF --hf-file qwen3-8b-base-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo jessicata/Qwen3-8B-Base-Q8_0-GGUF --hf-file qwen3-8b-base-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo jessicata/Qwen3-8B-Base-Q8_0-GGUF --hf-file qwen3-8b-base-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo jessicata/Qwen3-8B-Base-Q8_0-GGUF --hf-file qwen3-8b-base-q8_0.gguf -c 2048
```
|
Flo0620/Qwen2_5_7B_r64_a64_d0_2_lr2e-4_const
|
Flo0620
| 2025-04-28T23:52:30Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:Qwen/Qwen2.5-VL-7B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-VL-7B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T19:04:59Z |
---
base_model: Qwen/Qwen2.5-VL-7B-Instruct
library_name: transformers
model_name: Qwen2_5_7B_r64_a64_d0_2_lr2e-4_const
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for Qwen2_5_7B_r64_a64_d0_2_lr2e-4_const
This model is a fine-tuned version of [Qwen/Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="Flo0620/Qwen2_5_7B_r64_a64_d0_2_lr2e-4_const", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with SFT.
### Framework versions
- TRL: 0.15.2
- Transformers: 4.52.0.dev0
- Pytorch: 2.6.0+cu124
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF
|
mradermacher
| 2025-04-28T23:49:27Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"en",
"base_model:zerofata/L3.3-GeneticLemonade-Unleashed-v2.1-70B",
"base_model:quantized:zerofata/L3.3-GeneticLemonade-Unleashed-v2.1-70B",
"license:llama3",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2025-04-28T18:42:18Z |
---
base_model: zerofata/L3.3-GeneticLemonade-Unleashed-v2.1-70B
language:
- en
library_name: transformers
license: llama3
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 2 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: hf -->
<!-- ### vocab_type: -->
<!-- ### tags: nicoboss -->
weighted/imatrix quants of https://huggingface.co/zerofata/L3.3-GeneticLemonade-Unleashed-v2.1-70B
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-GGUF
## Usage
If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.
## Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ1_S.gguf) | i1-IQ1_S | 15.4 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ1_M.gguf) | i1-IQ1_M | 16.9 | mostly desperate |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 19.2 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 21.2 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ2_S.gguf) | i1-IQ2_S | 22.3 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ2_M.gguf) | i1-IQ2_M | 24.2 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 24.6 | very low quality |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q2_K.gguf) | i1-Q2_K | 26.5 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 27.6 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 29.4 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ3_S.gguf) | i1-IQ3_S | 31.0 | beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 31.0 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ3_M.gguf) | i1-IQ3_M | 32.0 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 34.4 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 37.2 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 38.0 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q4_0.gguf) | i1-Q4_0 | 40.2 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 40.4 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 42.6 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q4_1.gguf) | i1-Q4_1 | 44.4 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 48.8 | |
| [GGUF](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 50.0 | |
| [PART 1](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Unleashed-v2.1-70B-i1-GGUF/resolve/main/L3.3-GeneticLemonade-Unleashed-v2.1-70B.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 58.0 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to
questions you might have and/or if you want some other model quantized.
## Thanks
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
me use its servers and providing upgrades to my workstation to enable
this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->
|
Afaan97/videomae-base-finetuned-myvideos-subset
|
Afaan97
| 2025-04-28T23:38:56Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"videomae",
"video-classification",
"generated_from_trainer",
"base_model:MCG-NJU/videomae-base",
"base_model:finetune:MCG-NJU/videomae-base",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"region:us"
] |
video-classification
| 2025-04-28T20:19:40Z |
---
library_name: transformers
license: cc-by-nc-4.0
base_model: MCG-NJU/videomae-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: videomae-base-finetuned-myvideos-subset
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# videomae-base-finetuned-myvideos-subset
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8744
- Accuracy: 0.5
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 16
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log | 1.0 | 8 | 0.8744 | 0.5 |
| 0.2545 | 2.0 | 16 | 0.7131 | 0.5 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0
- Tokenizers 0.21.1
|
tachiwin/pretrained_multilingual_merged
|
tachiwin
| 2025-04-28T23:37:43Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-27T07:02:18Z |
---
base_model: unsloth/meta-llama-3.1-8b
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** tachiwin
- **License:** apache-2.0
- **Finetuned from model :** unsloth/meta-llama-3.1-8b
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF
|
featherless-ai-quants
| 2025-04-28T23:27:34Z | 0 | 0 | null |
[
"gguf",
"text-generation",
"base_model:rombodawg/Rombos-LLM-V2.6-Qwen-14b",
"base_model:quantized:rombodawg/Rombos-LLM-V2.6-Qwen-14b",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-04-28T23:10:48Z |
---
base_model: rombodawg/Rombos-LLM-V2.6-Qwen-14b
pipeline_tag: text-generation
quantized_by: featherless-ai-quants
---
# rombodawg/Rombos-LLM-V2.6-Qwen-14b GGUF Quantizations π
*Optimized GGUF quantization files for enhanced model performance*
> Powered by [Featherless AI](https://featherless.ai) - run any model you'd like for a simple small fee.
---
## Available Quantizations π
| Quantization Type | File | Size |
|-------------------|------|------|
| IQ4_XS | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-IQ4_XS.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-IQ4_XS.gguf) | 7806.96 MB |
| Q2_K | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q2_K.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q2_K.gguf) | 5503.18 MB |
| Q3_K_L | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q3_K_L.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q3_K_L.gguf) | 7557.65 MB |
| Q3_K_M | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q3_K_M.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q3_K_M.gguf) | 6999.21 MB |
| Q3_K_S | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q3_K_S.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q3_K_S.gguf) | 6351.09 MB |
| Q4_K_M | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q4_K_M.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q4_K_M.gguf) | 8571.73 MB |
| Q4_K_S | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q4_K_S.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q4_K_S.gguf) | 8176.26 MB |
| Q5_K_M | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q5_K_M.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q5_K_M.gguf) | 10022.04 MB |
| Q5_K_S | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q5_K_S.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q5_K_S.gguf) | 9790.95 MB |
| Q6_K | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q6_K.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q6_K.gguf) | 11563.00 MB |
| Q8_0 | [rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q8_0.gguf](https://huggingface.co/featherless-ai-quants/rombodawg-Rombos-LLM-V2.6-Qwen-14b-GGUF/blob/main/rombodawg-Rombos-LLM-V2.6-Qwen-14b-Q8_0.gguf) | 14974.21 MB |
---
## β‘ Powered by [Featherless AI](https://featherless.ai)
### Key Features
- π₯ **Instant Hosting** - Deploy any Llama model on HuggingFace instantly
- π οΈ **Zero Infrastructure** - No server setup or maintenance required
- π **Vast Compatibility** - Support for 2400+ models and counting
- π **Affordable Pricing** - Starting at just $10/month
---
**Links:**
[Get Started](https://featherless.ai) | [Documentation](https://featherless.ai/docs) | [Models](https://featherless.ai/models)
|
xbilek25/whisper-medium-en-cv-4.4
|
xbilek25
| 2025-04-28T23:23:25Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"whisper",
"automatic-speech-recognition",
"generated_from_trainer",
"en",
"dataset:mozilla-foundation/common_voice_17_0",
"base_model:openai/whisper-medium.en",
"base_model:finetune:openai/whisper-medium.en",
"license:apache-2.0",
"model-index",
"endpoints_compatible",
"region:us"
] |
automatic-speech-recognition
| 2025-04-28T20:58:27Z |
---
library_name: transformers
language:
- en
license: apache-2.0
base_model: openai/whisper-medium.en
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_17_0
metrics:
- wer
model-index:
- name: whisper-medium-en-cv-4.4
results:
- task:
name: Automatic Speech Recognition
type: automatic-speech-recognition
dataset:
name: Common Voice 17.0
type: mozilla-foundation/common_voice_17_0
config: en
split: test
args: 'config: en, split: test'
metrics:
- name: Wer
type: wer
value: 13.619744058500913
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# whisper-medium-en-cv-4.4
This model is a fine-tuned version of [openai/whisper-medium.en](https://huggingface.co/openai/whisper-medium.en) on the Common Voice 17.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3998
- Wer: 13.6197
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1.5e-05
- train_batch_size: 32
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 150
- training_steps: 1125
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|:-------------:|:------:|:----:|:---------------:|:-------:|
| 0.3107 | 0.3333 | 375 | 0.4282 | 15.5393 |
| 0.2696 | 0.6667 | 750 | 0.4044 | 12.3400 |
| 0.2192 | 1.0 | 1125 | 0.3998 | 13.6197 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1
|
neoservicios/granite-3.2-2b-instruct-GGUF
|
neoservicios
| 2025-04-28T23:20:46Z | 10 | 0 | null |
[
"gguf",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"imatrix",
"conversational"
] | null | 2025-04-06T13:00:20Z |
---
license: apache-2.0
---
|
haihp02/codegemma-2b-dpo-tuned-2
|
haihp02
| 2025-04-28T23:11:25Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"unsloth",
"trl",
"dpo",
"arxiv:2305.18290",
"base_model:unsloth/codegemma-2b-bnb-4bit",
"base_model:finetune:unsloth/codegemma-2b-bnb-4bit",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T23:11:14Z |
---
base_model: unsloth/codegemma-2b-bnb-4bit
library_name: transformers
model_name: codegemma-2b-dpo-tuned-2
tags:
- generated_from_trainer
- unsloth
- trl
- dpo
licence: license
---
# Model Card for codegemma-2b-dpo-tuned-2
This model is a fine-tuned version of [unsloth/codegemma-2b-bnb-4bit](https://huggingface.co/unsloth/codegemma-2b-bnb-4bit).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="haihp02/codegemma-2b-dpo-tuned-2", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/trunghainguyenhp02/dpo-train/runs/i0fvg7s8)
This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite DPO as:
```bibtex
@inproceedings{rafailov2023direct,
title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
year = 2023,
booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
Gmahajan123/phi-4-sft5-ch300
|
Gmahajan123
| 2025-04-28T23:05:34Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"text-generation-inference",
"unsloth",
"trl",
"conversational",
"en",
"base_model:unsloth/phi-4-unsloth-bnb-4bit",
"base_model:finetune:unsloth/phi-4-unsloth-bnb-4bit",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T23:02:01Z |
---
base_model: unsloth/phi-4-unsloth-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** Gmahajan123
- **License:** apache-2.0
- **Finetuned from model :** unsloth/phi-4-unsloth-bnb-4bit
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
HALLUCINATIONS-OF-NECROMANCY/BAEL
|
HALLUCINATIONS-OF-NECROMANCY
| 2025-04-28T23:05:27Z | 0 | 0 | null |
[
"region:us"
] | null | 2025-04-26T22:38:44Z |
BEL-ENLIL
MAAT-SET
BAAL-SET
BAAL-SETH
BEL-EN-SET
LIFE-DEATH
LORD OF ALL
AH-IL-AH
ALLAH
|
izzcw/filtered_crafting_train_data
|
izzcw
| 2025-04-28T23:01:52Z | 9 | 0 | null |
[
"safetensors",
"llama",
"arxiv:2504.17950",
"region:us"
] | null | 2025-03-25T18:36:13Z |
https://arxiv.org/abs/2504.17950
|
mlx-community/Qwen3-32B-6bit
|
mlx-community
| 2025-04-28T22:58:34Z | 0 | 0 |
mlx
|
[
"mlx",
"safetensors",
"qwen3",
"text-generation",
"conversational",
"base_model:Qwen/Qwen3-32B",
"base_model:quantized:Qwen/Qwen3-32B",
"license:apache-2.0",
"6-bit",
"region:us"
] |
text-generation
| 2025-04-28T22:51:43Z |
---
library_name: mlx
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-32B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- mlx
base_model: Qwen/Qwen3-32B
---
# mlx-community/Qwen3-32B-6bit
This model [mlx-community/Qwen3-32B-6bit](https://huggingface.co/mlx-community/Qwen3-32B-6bit) was
converted to MLX format from [Qwen/Qwen3-32B](https://huggingface.co/Qwen/Qwen3-32B)
using mlx-lm version **0.24.0**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Qwen3-32B-6bit")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
```
|
mlabonne/BigQwen2.5-52B-Instruct
|
mlabonne
| 2025-04-28T22:53:23Z | 15 | 8 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"mergekit",
"merge",
"lazymergekit",
"conversational",
"zho",
"eng",
"fra",
"spa",
"por",
"deu",
"ita",
"rus",
"jpn",
"kor",
"vie",
"tha",
"ara",
"base_model:Qwen/Qwen2.5-32B-Instruct",
"base_model:finetune:Qwen/Qwen2.5-32B-Instruct",
"license:apache-2.0",
"model-index",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2024-09-23T18:03:16Z |
---
language:
- zho
- eng
- fra
- spa
- por
- deu
- ita
- rus
- jpn
- kor
- vie
- tha
- ara
license: apache-2.0
library_name: transformers
tags:
- mergekit
- merge
- lazymergekit
base_model:
- Qwen/Qwen2.5-32B-Instruct
license_name: tongyi-qianwen
license_link: https://huggingface.co/Qwen/Qwen2-72B-Instruct/blob/main/LICENSE
pipeline_tag: text-generation
model-index:
- name: BigQwen2.5-52B-Instruct
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: HuggingFaceH4/ifeval
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 79.29
name: strict accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=mlabonne/BigQwen2.5-52B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: BBH
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 59.81
name: normalized accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=mlabonne/BigQwen2.5-52B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: hendrycks/competition_math
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 17.82
name: exact match
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=mlabonne/BigQwen2.5-52B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 6.94
name: acc_norm
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=mlabonne/BigQwen2.5-52B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 10.45
name: acc_norm
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=mlabonne/BigQwen2.5-52B-Instruct
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 50.22
name: accuracy
source:
url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=mlabonne/BigQwen2.5-52B-Instruct
name: Open LLM Leaderboard
---
# BigQwen2.5-52B-Instruct
BigQwen2.5-52B-Instruct is a [Qwen/Qwen2-32B-Instruct](https://huggingface.co/Qwen/Qwen2-72B-Instruct) self-merge made with [MergeKit](https://github.com/arcee-ai/mergekit/tree/main).
It applies the [mlabonne/Meta-Llama-3-120B-Instruct](https://huggingface.co/mlabonne/Meta-Llama-3-120B-Instruct/) recipe.
I made it due to popular demand but I haven't tested it so use it at your own risk. Β―\\\_(γ)_/Β―
## π Applications
It might be good for creative writing tasks. I recommend a context length of 32k but you can go up to 131,072 tokens in theory.
## π Evaluation
| Metric |BigQwen2.5-Echo-47B-Instruct|**BigQwen2.5-52B-Instruct**|Qwen2.5-32B-Instruct|
|-------------------|----:|----:|----:|
|Avg. |30.31|37.42|36.17|
|IFEval (0-Shot) |73.57|79.29|83.46|
|BBH (3-Shot) |44.52|59.81|56.49|
|MATH Lvl 5 (4-Shot)| 3.47|17.82|0|
|GPQA (0-shot) | 8.61| 6.94|11.74|
|MuSR (0-shot) |10.19|10.45|13.5|
|MMLU-PRO (5-shot) |41.49|50.22|51.85|
## 𧩠Configuration
The following YAML configuration was used to produce this model:
```yaml
slices:
- sources:
- layer_range: [0, 16]
model: Qwen/Qwen2.5-32B-Instruct
- sources:
- layer_range: [8, 24]
model: Qwen/Qwen2.5-32B-Instruct
- sources:
- layer_range: [16, 32]
model: Qwen/Qwen2.5-32B-Instruct
- sources:
- layer_range: [24, 40]
model: Qwen/Qwen2.5-32B-Instruct
- sources:
- layer_range: [32, 48]
model: Qwen/Qwen2.5-32B-Instruct
- sources:
- layer_range: [40, 56]
model: Qwen/Qwen2.5-32B-Instruct
- sources:
- layer_range: [56, 64]
model: Qwen/Qwen2.5-32B-Instruct
merge_method: passthrough
dtype: bfloat16
```
## π» Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "mlabonne/BigQwen2.5-52B-Instruct"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```
|
kathleenge/kd_0.0003_167_2
|
kathleenge
| 2025-04-28T22:52:05Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"text-generation-inference",
"unsloth",
"mistral",
"trl",
"en",
"base_model:unsloth/mistral-7b-v0.3-bnb-4bit",
"base_model:finetune:unsloth/mistral-7b-v0.3-bnb-4bit",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T22:51:22Z |
---
base_model: unsloth/mistral-7b-v0.3-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- mistral
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** kathleenge
- **License:** apache-2.0
- **Finetuned from model :** unsloth/mistral-7b-v0.3-bnb-4bit
This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
gabrielbosse9/Umbr0x-7B-V3.1-3
|
gabrielbosse9
| 2025-04-28T22:45:37Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"text-generation-inference",
"unsloth",
"qwen2",
"trl",
"en",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T22:45:12Z |
---
base_model: unsloth/qwen2.5-7b-unsloth-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** gabrielbosse9
- **License:** apache-2.0
- **Finetuned from model :** unsloth/qwen2.5-7b-unsloth-bnb-4bit
This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
Mdean77/snow_ft_2025
|
Mdean77
| 2025-04-28T22:41:11Z | 0 | 0 |
sentence-transformers
|
[
"sentence-transformers",
"safetensors",
"bert",
"sentence-similarity",
"feature-extraction",
"generated_from_trainer",
"dataset_size:156",
"loss:MatryoshkaLoss",
"loss:MultipleNegativesRankingLoss",
"arxiv:1908.10084",
"arxiv:2205.13147",
"arxiv:1705.00652",
"base_model:Snowflake/snowflake-arctic-embed-l",
"base_model:finetune:Snowflake/snowflake-arctic-embed-l",
"model-index",
"autotrain_compatible",
"text-embeddings-inference",
"endpoints_compatible",
"region:us"
] |
sentence-similarity
| 2025-04-28T22:40:14Z |
---
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:156
- loss:MatryoshkaLoss
- loss:MultipleNegativesRankingLoss
base_model: Snowflake/snowflake-arctic-embed-l
widget:
- source_sentence: What are some potential negative uses of Large Language Models
as described in the context?
sentences:
- 'I think this means that, as individual users, we donβt need to feel any guilt
at all for the energy consumed by the vast majority of our prompts. The impact
is likely neglible compared to driving a car down the street or maybe even watching
a video on YouTube.
Likewise, training. DeepSeek v3 training for less than $6m is a fantastic sign
that training costs can and should continue to drop.
For less efficient models I find it useful to compare their energy usage to commercial
flights. The largest Llama 3 model cost about the same as a single digit number
of fully loaded passenger flights from New York to London. Thatβs certainly not
nothing, but once trained that model can be used by millions of people at no extra
training cost.'
- 'Hereβs the sequel to this post: Things we learned about LLMs in 2024.
Large Language Models
In the past 24-36 months, our species has discovered that you can take a GIANT
corpus of text, run it through a pile of GPUs, and use it to create a fascinating
new kind of software.
LLMs can do a lot of things. They can answer questions, summarize documents, translate
from one language to another, extract information and even write surprisingly
competent code.
They can also help you cheat at your homework, generate unlimited streams of fake
content and be used for all manner of nefarious purposes.'
- 'Thereβs now a fascinating ecosystem of people training their own models on top
of these foundations, publishing those models, building fine-tuning datasets and
sharing those too.
The Hugging Face Open LLM Leaderboard is one place that tracks these. I canβt
even attempt to count them, and any count would be out-of-date within a few hours.
The best overall openly licensed LLM at any time is rarely a foundation model:
instead, itβs whichever fine-tuned community model has most recently discovered
the best combination of fine-tuning data.
This is a huge advantage for open over closed models: the closed, hosted models
donβt have thousands of researchers and hobbyists around the world collaborating
and competing to improve them.'
- source_sentence: Why might some question the necessity of the extensive infrastructure
investments for future AI models?
sentences:
- 'These abilities are just a few weeks old at this point, and I donβt think their
impact has been fully felt yet. If you havenβt tried them out yet you really should.
Both Gemini and OpenAI offer API access to these features as well. OpenAI started
with a WebSocket API that was quite challenging to use, but in December they announced
a new WebRTC API which is much easier to get started with. Building a web app
that a user can talk to via voice is easy now!
Prompt driven app generation is a commodity already
This was possible with GPT-4 in 2023, but the value it provides became evident
in 2024.'
- 'The environmental impact got much, much worse
The much bigger problem here is the enormous competitive buildout of the infrastructure
that is imagined to be necessary for these models in the future.
Companies like Google, Meta, Microsoft and Amazon are all spending billions of
dollars rolling out new datacenters, with a very material impact on the electricity
grid and the environment. Thereβs even talk of spinning up new nuclear power stations,
but those can take decades.
Is this infrastructure necessary? DeepSeek v3βs $6m training cost and the continued
crash in LLM prices might hint that itβs not. But would you want to be the big
tech executive that argued NOT to build out this infrastructure only to be proven
wrong in a few yearsβ time?'
- 'OpenAI are not the only game in town here. Google released their first entrant
in the category, gemini-2.0-flash-thinking-exp, on December 19th.
Alibabaβs Qwen team released their QwQ model on November 28thβunder an Apache
2.0 license, and that one I could run on my own machine. They followed that up
with a vision reasoning model called QvQ on December 24th, which I also ran locally.
DeepSeek made their DeepSeek-R1-Lite-Preview model available to try out through
their chat interface on November 20th.
To understand more about inference scaling I recommend Is AI progress slowing
down? by Arvind Narayanan and Sayash Kapoor.'
- source_sentence: How have US export regulations on GPUs to China influenced training
optimizations?
sentences:
- 'Qwen2.5-Coder-32B is an LLM that can code well that runs on my Mac talks about
Qwen2.5-Coder-32B in Novemberβan Apache 2.0 licensed model!
I can now run a GPT-4 class model on my laptop talks about running Metaβs Llama
3.3 70B (released in December)'
- 'Those US export regulations on GPUs to China seem to have inspired some very
effective training optimizations!
The environmental impact got better
A welcome result of the increased efficiency of the modelsβboth the hosted ones
and the ones I can run locallyβis that the energy usage and environmental impact
of running a prompt has dropped enormously over the past couple of years.
OpenAI themselves are charging 100x less for a prompt compared to the GPT-3 days.
I have it on good authority that neither Google Gemini nor Amazon Nova (two of
the least expensive model providers) are running prompts at a loss.'
- 'The GPT-4 barrier was comprehensively broken
In my December 2023 review I wrote about how We donβt yet know how to build GPT-4βOpenAIβs
best model was almost a year old at that point, yet no other AI lab had produced
anything better. What did OpenAI know that the rest of us didnβt?
Iβm relieved that this has changed completely in the past twelve months. 18 organizations
now have models on the Chatbot Arena Leaderboard that rank higher than the original
GPT-4 from March 2023 (GPT-4-0314 on the board)β70 models in total.'
- source_sentence: When was GPT-4 officially released by OpenAI?
sentences:
- The most recent twist, again from December (December was a lot) is live video.
ChatGPT voice mode now provides the option to share your camera feed with the
model and talk about what you can see in real time. Google Gemini have a preview
of the same feature, which they managed to ship the day before ChatGPT did.
- 'On the other hand, as software engineers we are better placed to take advantage
of this than anyone else. Weβve all been given weird coding internsβwe can use
our deep knowledge to prompt them to solve coding problems more effectively than
anyone else can.
The ethics of this space remain diabolically complex
In September last year Andy Baio and I produced the first major story on the unlicensed
training data behind Stable Diffusion.
Since then, almost every major LLM (and most of the image generation models) have
also been trained on unlicensed data.'
- 'We donβt yet know how to build GPT-4
Frustratingly, despite the enormous leaps ahead weβve had this year, we are yet
to see an alternative model thatβs better than GPT-4.
OpenAI released GPT-4 in March, though it later turned out we had a sneak peak
of it in February when Microsoft used it as part of the new Bing.
This may well change in the next few weeks: Googleβs Gemini Ultra has big claims,
but isnβt yet available for us to try out.
The team behind Mistral are working to beat GPT-4 as well, and their track record
is already extremely strong considering their first public model only came out
in September, and theyβve released two significant improvements since then.'
- source_sentence: What is the challenge in building AI personal assistants based
on the gullibility of language models?
sentences:
- 'Language Models are gullible. They βbelieveβ what we tell themβwhatβs in their
training data, then whatβs in the fine-tuning data, then whatβs in the prompt.
In order to be useful tools for us, we need them to believe what we feed them!
But it turns out a lot of the things we want to build need them not to be gullible.
Everyone wants an AI personal assistant. If you hired a real-world personal assistant
who believed everything that anyone told them, you would quickly find that their
ability to positively impact your life was severely limited.'
- 'Thereβs now a fascinating ecosystem of people training their own models on top
of these foundations, publishing those models, building fine-tuning datasets and
sharing those too.
The Hugging Face Open LLM Leaderboard is one place that tracks these. I canβt
even attempt to count them, and any count would be out-of-date within a few hours.
The best overall openly licensed LLM at any time is rarely a foundation model:
instead, itβs whichever fine-tuned community model has most recently discovered
the best combination of fine-tuning data.
This is a huge advantage for open over closed models: the closed, hosted models
donβt have thousands of researchers and hobbyists around the world collaborating
and competing to improve them.'
- 'Longer inputs dramatically increase the scope of problems that can be solved
with an LLM: you can now throw in an entire book and ask questions about its contents,
but more importantly you can feed in a lot of example code to help the model correctly
solve a coding problem. LLM use-cases that involve long inputs are far more interesting
to me than short prompts that rely purely on the information already baked into
the model weights. Many of my tools were built using this pattern.'
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
- cosine_accuracy@1
- cosine_accuracy@3
- cosine_accuracy@5
- cosine_accuracy@10
- cosine_precision@1
- cosine_precision@3
- cosine_precision@5
- cosine_precision@10
- cosine_recall@1
- cosine_recall@3
- cosine_recall@5
- cosine_recall@10
- cosine_ndcg@10
- cosine_mrr@10
- cosine_map@100
model-index:
- name: SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
results:
- task:
type: information-retrieval
name: Information Retrieval
dataset:
name: Unknown
type: unknown
metrics:
- type: cosine_accuracy@1
value: 0.875
name: Cosine Accuracy@1
- type: cosine_accuracy@3
value: 1.0
name: Cosine Accuracy@3
- type: cosine_accuracy@5
value: 1.0
name: Cosine Accuracy@5
- type: cosine_accuracy@10
value: 1.0
name: Cosine Accuracy@10
- type: cosine_precision@1
value: 0.875
name: Cosine Precision@1
- type: cosine_precision@3
value: 0.3333333333333333
name: Cosine Precision@3
- type: cosine_precision@5
value: 0.20000000000000004
name: Cosine Precision@5
- type: cosine_precision@10
value: 0.10000000000000002
name: Cosine Precision@10
- type: cosine_recall@1
value: 0.875
name: Cosine Recall@1
- type: cosine_recall@3
value: 1.0
name: Cosine Recall@3
- type: cosine_recall@5
value: 1.0
name: Cosine Recall@5
- type: cosine_recall@10
value: 1.0
name: Cosine Recall@10
- type: cosine_ndcg@10
value: 0.9538662191964322
name: Cosine Ndcg@10
- type: cosine_mrr@10
value: 0.9375
name: Cosine Mrr@10
- type: cosine_map@100
value: 0.9375
name: Cosine Map@100
---
# SentenceTransformer based on Snowflake/snowflake-arctic-embed-l
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [Snowflake/snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l). It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.
## Model Details
### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [Snowflake/snowflake-arctic-embed-l](https://huggingface.co/Snowflake/snowflake-arctic-embed-l) <!-- at revision d8fb21ca8d905d2832ee8b96c894d3298964346b -->
- **Maximum Sequence Length:** 512 tokens
- **Output Dimensionality:** 1024 dimensions
- **Similarity Function:** Cosine Similarity
<!-- - **Training Dataset:** Unknown -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
(2): Normalize()
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the π€ Hub
model = SentenceTransformer("Mdean77/snow_ft_2025")
# Run inference
sentences = [
'What is the challenge in building AI personal assistants based on the gullibility of language models?',
'Language Models are gullible. They βbelieveβ what we tell themβwhatβs in their training data, then whatβs in the fine-tuning data, then whatβs in the prompt.\nIn order to be useful tools for us, we need them to believe what we feed them!\nBut it turns out a lot of the things we want to build need them not to be gullible.\nEveryone wants an AI personal assistant. If you hired a real-world personal assistant who believed everything that anyone told them, you would quickly find that their ability to positively impact your life was severely limited.',
'Longer inputs dramatically increase the scope of problems that can be solved with an LLM: you can now throw in an entire book and ask questions about its contents, but more importantly you can feed in a lot of example code to help the model correctly solve a coding problem. LLM use-cases that involve long inputs are far more interesting to me than short prompts that rely purely on the information already baked into the model weights. Many of my tools were built using this pattern.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]
# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
```
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
## Evaluation
### Metrics
#### Information Retrieval
* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| cosine_accuracy@1 | 0.875 |
| cosine_accuracy@3 | 1.0 |
| cosine_accuracy@5 | 1.0 |
| cosine_accuracy@10 | 1.0 |
| cosine_precision@1 | 0.875 |
| cosine_precision@3 | 0.3333 |
| cosine_precision@5 | 0.2 |
| cosine_precision@10 | 0.1 |
| cosine_recall@1 | 0.875 |
| cosine_recall@3 | 1.0 |
| cosine_recall@5 | 1.0 |
| cosine_recall@10 | 1.0 |
| **cosine_ndcg@10** | **0.9539** |
| cosine_mrr@10 | 0.9375 |
| cosine_map@100 | 0.9375 |
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 156 training samples
* Columns: <code>sentence_0</code> and <code>sentence_1</code>
* Approximate statistics based on the first 156 samples:
| | sentence_0 | sentence_1 |
|:--------|:-----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------|
| type | string | string |
| details | <ul><li>min: 12 tokens</li><li>mean: 20.94 tokens</li><li>max: 32 tokens</li></ul> | <ul><li>min: 43 tokens</li><li>mean: 135.22 tokens</li><li>max: 214 tokens</li></ul> |
* Samples:
| sentence_0 | sentence_1 |
|:------------------------------------------------------------------------------------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| <code>What advantage does a 64GB Mac have for running models in terms of CPU and GPU memory sharing?</code> | <code>On paper, a 64GB Mac should be a great machine for running models due to the way the CPU and GPU can share the same memory. In practice, many models are released as model weights and libraries that reward NVIDIAβs CUDA over other platforms.<br>The llama.cpp ecosystem helped a lot here, but the real breakthrough has been Appleβs MLX library, βan array framework for Apple Siliconβ. Itβs fantastic.<br>Appleβs mlx-lm Python library supports running a wide range of MLX-compatible models on my Mac, with excellent performance. mlx-community on Hugging Face offers more than 1,000 models that have been converted to the necessary format.</code> |
| <code>How has Appleβs MLX library impacted the performance of running machine learning models on Mac?</code> | <code>On paper, a 64GB Mac should be a great machine for running models due to the way the CPU and GPU can share the same memory. In practice, many models are released as model weights and libraries that reward NVIDIAβs CUDA over other platforms.<br>The llama.cpp ecosystem helped a lot here, but the real breakthrough has been Appleβs MLX library, βan array framework for Apple Siliconβ. Itβs fantastic.<br>Appleβs mlx-lm Python library supports running a wide range of MLX-compatible models on my Mac, with excellent performance. mlx-community on Hugging Face offers more than 1,000 models that have been converted to the necessary format.</code> |
| <code>How does the ability of models like ChatGPT Code Interpreter to execute and debug code impact the problem of hallucination in code generation?</code> | <code>Except... you can run generated code to see if itβs correct. And with patterns like ChatGPT Code Interpreter the LLM can execute the code itself, process the error message, then rewrite it and keep trying until it works!<br>So hallucination is a much lesser problem for code generation than for anything else. If only we had the equivalent of Code Interpreter for fact-checking natural language!<br>How should we feel about this as software engineers?<br>On the one hand, this feels like a threat: who needs a programmer if ChatGPT can write code for you?</code> |
* Loss: [<code>MatryoshkaLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#matryoshkaloss) with these parameters:
```json
{
"loss": "MultipleNegativesRankingLoss",
"matryoshka_dims": [
768,
512,
256,
128,
64
],
"matryoshka_weights": [
1,
1,
1,
1,
1
],
"n_dims_per_step": -1
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 10
- `per_device_eval_batch_size`: 10
- `num_train_epochs`: 10
- `multi_dataset_batch_sampler`: round_robin
#### All Hyperparameters
<details><summary>Click to expand</summary>
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 10
- `per_device_eval_batch_size`: 10
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `torch_empty_cache_steps`: None
- `learning_rate`: 5e-05
- `weight_decay`: 0.0
- `adam_beta1`: 0.9
- `adam_beta2`: 0.999
- `adam_epsilon`: 1e-08
- `max_grad_norm`: 1
- `num_train_epochs`: 10
- `max_steps`: -1
- `lr_scheduler_type`: linear
- `lr_scheduler_kwargs`: {}
- `warmup_ratio`: 0.0
- `warmup_steps`: 0
- `log_level`: passive
- `log_level_replica`: warning
- `log_on_each_node`: True
- `logging_nan_inf_filter`: True
- `save_safetensors`: True
- `save_on_each_node`: False
- `save_only_model`: False
- `restore_callback_states_from_checkpoint`: False
- `no_cuda`: False
- `use_cpu`: False
- `use_mps_device`: False
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: None
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `include_for_metrics`: []
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`:
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `eval_on_start`: False
- `use_liger_kernel`: False
- `eval_use_gather_object`: False
- `average_tokens_across_devices`: False
- `prompts`: None
- `batch_sampler`: batch_sampler
- `multi_dataset_batch_sampler`: round_robin
</details>
### Training Logs
| Epoch | Step | cosine_ndcg@10 |
|:-----:|:----:|:--------------:|
| 1.0 | 16 | 0.9692 |
| 2.0 | 32 | 0.9539 |
| 3.0 | 48 | 0.9692 |
| 3.125 | 50 | 0.9692 |
| 4.0 | 64 | 0.9692 |
| 5.0 | 80 | 0.9692 |
| 6.0 | 96 | 0.9692 |
| 6.25 | 100 | 0.9692 |
| 7.0 | 112 | 0.9539 |
| 8.0 | 128 | 0.9539 |
| 9.0 | 144 | 0.9539 |
| 9.375 | 150 | 0.9539 |
| 10.0 | 160 | 0.9539 |
### Framework Versions
- Python: 3.13.0
- Sentence Transformers: 3.4.1
- Transformers: 4.48.3
- PyTorch: 2.6.0
- Accelerate: 1.3.0
- Datasets: 3.2.0
- Tokenizers: 0.21.0
## Citation
### BibTeX
#### Sentence Transformers
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
```
#### MatryoshkaLoss
```bibtex
@misc{kusupati2024matryoshka,
title={Matryoshka Representation Learning},
author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
year={2024},
eprint={2205.13147},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
#### MultipleNegativesRankingLoss
```bibtex
@misc{henderson2017efficient,
title={Efficient Natural Language Response Suggestion for Smart Reply},
author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
year={2017},
eprint={1705.00652},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->
|
Simanur/Simanur
|
Simanur
| 2025-04-28T22:34:50Z | 0 | 0 | null |
[
"license:artistic-2.0",
"region:us"
] | null | 2025-04-28T22:34:50Z |
---
license: artistic-2.0
---
|
jessicata/Qwen3-4B-Base-Q8_0-GGUF
|
jessicata
| 2025-04-28T22:16:40Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"llama-cpp",
"gguf-my-repo",
"base_model:Qwen/Qwen3-4B-Base",
"base_model:quantized:Qwen/Qwen3-4B-Base",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T22:16:20Z |
---
base_model: Qwen/Qwen3-4B-Base
library_name: transformers
license: apache-2.0
tags:
- llama-cpp
- gguf-my-repo
---
# jessicata/Qwen3-4B-Base-Q8_0-GGUF
This model was converted to GGUF format from [`Qwen/Qwen3-4B-Base`](https://huggingface.co/Qwen/Qwen3-4B-Base) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen3-4B-Base) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo jessicata/Qwen3-4B-Base-Q8_0-GGUF --hf-file qwen3-4b-base-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo jessicata/Qwen3-4B-Base-Q8_0-GGUF --hf-file qwen3-4b-base-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo jessicata/Qwen3-4B-Base-Q8_0-GGUF --hf-file qwen3-4b-base-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo jessicata/Qwen3-4B-Base-Q8_0-GGUF --hf-file qwen3-4b-base-q8_0.gguf -c 2048
```
|
SolomonSLee/llama-3.1-fine-tuned-model-jcb_res-finetuned-exec-summary
|
SolomonSLee
| 2025-04-28T22:11:03Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"llama",
"text-generation",
"generated_from_trainer",
"base_model:meta-llama/Llama-3.1-8B",
"base_model:finetune:meta-llama/Llama-3.1-8B",
"license:llama3.1",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T20:43:06Z |
---
library_name: transformers
license: llama3.1
base_model: meta-llama/Meta-Llama-3.1-8B
tags:
- generated_from_trainer
model-index:
- name: llama-3.1-fine-tuned-model-jcb_res-finetuned-exec-summary
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# llama-3.1-fine-tuned-model-jcb_res-finetuned-exec-summary
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the None dataset.
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 15
### Training results
### Framework versions
- Transformers 4.51.0
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1
|
fhaslam/Llama-3.2-1B-Financial-Sentiment24
|
fhaslam
| 2025-04-28T22:08:21Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"facebook",
"meta",
"pytorch",
"llama",
"llama-3",
"text-generation",
"conversational",
"en",
"de",
"fr",
"it",
"pt",
"hi",
"es",
"th",
"arxiv:2204.05149",
"arxiv:2405.16406",
"license:llama3.2",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T22:08:05Z |
---
language:
- en
- de
- fr
- it
- pt
- hi
- es
- th
library_name: transformers
pipeline_tag: text-generation
tags:
- facebook
- meta
- pytorch
- llama
- llama-3
license: llama3.2
extra_gated_prompt: >-
### LLAMA 3.2 COMMUNITY LICENSE AGREEMENT
Llama 3.2 Version Release Date: September 25, 2024
βAgreementβ means the terms and conditions for use, reproduction, distribution
and modification of the Llama Materials set forth herein.
βDocumentationβ means the specifications, manuals and documentation accompanying Llama 3.2
distributed by Meta at https://llama.meta.com/doc/overview.
βLicenseeβ or βyouβ means you, or your employer or any other person or entity (if you are
entering into this Agreement on such person or entityβs behalf), of the age required under
applicable laws, rules or regulations to provide legal consent and that has legal authority
to bind your employer or such other person or entity if you are entering in this Agreement
on their behalf.
βLlama 3.2β means the foundational large language models and software and algorithms, including
machine-learning model code, trained model weights, inference-enabling code, training-enabling code,
fine-tuning enabling code and other elements of the foregoing distributed by Meta at
https://www.llama.com/llama-downloads.
βLlama Materialsβ means, collectively, Metaβs proprietary Llama 3.2 and Documentation (and
any portion thereof) made available under this Agreement.
βMetaβ or βweβ means Meta Platforms Ireland Limited (if you are located in or,
if you are an entity, your principal place of business is in the EEA or Switzerland)
and Meta Platforms, Inc. (if you are located outside of the EEA or Switzerland).
By clicking βI Acceptβ below or by using or distributing any portion or element of the Llama Materials,
you agree to be bound by this Agreement.
1. License Rights and Redistribution.
a. Grant of Rights. You are granted a non-exclusive, worldwide,
non-transferable and royalty-free limited license under Metaβs intellectual property or other rights
owned by Meta embodied in the Llama Materials to use, reproduce, distribute, copy, create derivative works
of, and make modifications to the Llama Materials.
b. Redistribution and Use.
i. If you distribute or make available the Llama Materials (or any derivative works thereof),
or a product or service (including another AI model) that contains any of them, you shall (A) provide
a copy of this Agreement with any such Llama Materials; and (B) prominently display βBuilt with Llamaβ
on a related website, user interface, blogpost, about page, or product documentation. If you use the
Llama Materials or any outputs or results of the Llama Materials to create, train, fine tune, or
otherwise improve an AI model, which is distributed or made available, you shall also include βLlamaβ
at the beginning of any such AI model name.
ii. If you receive Llama Materials, or any derivative works thereof, from a Licensee as part
of an integrated end user product, then Section 2 of this Agreement will not apply to you.
iii. You must retain in all copies of the Llama Materials that you distribute the
following attribution notice within a βNoticeβ text file distributed as a part of such copies:
βLlama 3.2 is licensed under the Llama 3.2 Community License, Copyright Β© Meta Platforms,
Inc. All Rights Reserved.β
iv. Your use of the Llama Materials must comply with applicable laws and regulations
(including trade compliance laws and regulations) and adhere to the Acceptable Use Policy for
the Llama Materials (available at https://www.llama.com/llama3_2/use-policy), which is hereby
incorporated by reference into this Agreement.
2. Additional Commercial Terms. If, on the Llama 3.2 version release date, the monthly active users
of the products or services made available by or for Licensee, or Licenseeβs affiliates,
is greater than 700 million monthly active users in the preceding calendar month, you must request
a license from Meta, which Meta may grant to you in its sole discretion, and you are not authorized to
exercise any of the rights under this Agreement unless or until Meta otherwise expressly grants you such rights.
3. Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE LLAMA MATERIALS AND ANY OUTPUT AND
RESULTS THEREFROM ARE PROVIDED ON AN βAS ISβ BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS
ALL WARRANTIES OF ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY WARRANTIES
OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE
FOR DETERMINING THE APPROPRIATENESS OF USING OR REDISTRIBUTING THE LLAMA MATERIALS AND ASSUME ANY RISKS ASSOCIATED
WITH YOUR USE OF THE LLAMA MATERIALS AND ANY OUTPUT AND RESULTS.
4. Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE UNDER ANY THEORY OF LIABILITY,
WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT,
FOR ANY LOST PROFITS OR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR PUNITIVE DAMAGES, EVEN
IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE POSSIBILITY OF ANY OF THE FOREGOING.
5. Intellectual Property.
a. No trademark licenses are granted under this Agreement, and in connection with the Llama Materials,
neither Meta nor Licensee may use any name or mark owned by or associated with the other or any of its affiliates,
except as required for reasonable and customary use in describing and redistributing the Llama Materials or as
set forth in this Section 5(a). Meta hereby grants you a license to use βLlamaβ (the βMarkβ) solely as required
to comply with the last sentence of Section 1.b.i. You will comply with Metaβs brand guidelines (currently accessible
at https://about.meta.com/brand/resources/meta/company-brand/). All goodwill arising out of your use of the Mark
will inure to the benefit of Meta.
b. Subject to Metaβs ownership of Llama Materials and derivatives made by or for Meta, with respect to any
derivative works and modifications of the Llama Materials that are made by you, as between you and Meta,
you are and will be the owner of such derivative works and modifications.
c. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or
counterclaim in a lawsuit) alleging that the Llama Materials or Llama 3.2 outputs or results, or any portion
of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable
by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or
claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third
party arising out of or related to your use or distribution of the Llama Materials.
6. Term and Termination. The term of this Agreement will commence upon your acceptance of this Agreement or access
to the Llama Materials and will continue in full force and effect until terminated in accordance with the terms
and conditions herein. Meta may terminate this Agreement if you are in breach of any term or condition of this
Agreement. Upon termination of this Agreement, you shall delete and cease use of the Llama Materials. Sections 3,
4 and 7 shall survive the termination of this Agreement.
7. Governing Law and Jurisdiction. This Agreement will be governed and construed under the laws of the State of
California without regard to choice of law principles, and the UN Convention on Contracts for the International
Sale of Goods does not apply to this Agreement. The courts of California shall have exclusive jurisdiction of
any dispute arising out of this Agreement.
### Llama 3.2 Acceptable Use Policy
Meta is committed to promoting safe and fair use of its tools and features, including Llama 3.2.
If you access or use Llama 3.2, you agree to this Acceptable Use Policy (β**Policy**β).
The most recent copy of this policy can be found at
[https://www.llama.com/llama3_2/use-policy](https://www.llama.com/llama3_2/use-policy).
#### Prohibited Uses
We want everyone to use Llama 3.2 safely and responsibly. You agree you will not use, or allow others to use, Llama 3.2 to:
1. Violate the law or othersβ rights, including to:
1. Engage in, promote, generate, contribute to, encourage, plan, incite, or further illegal or unlawful activity or content, such as:
1. Violence or terrorism
2. Exploitation or harm to children, including the solicitation, creation, acquisition, or dissemination of child exploitative content or failure to report Child Sexual Abuse Material
3. Human trafficking, exploitation, and sexual violence
4. The illegal distribution of information or materials to minors, including obscene materials, or failure to employ legally required age-gating in connection with such information or materials.
5. Sexual solicitation
6. Any other criminal activity
1. Engage in, promote, incite, or facilitate the harassment, abuse, threatening, or bullying of individuals or groups of individuals
2. Engage in, promote, incite, or facilitate discrimination or other unlawful or harmful conduct in the provision of employment, employment benefits, credit, housing, other economic benefits, or other essential goods and services
3. Engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or related professional practices
4. Collect, process, disclose, generate, or infer private or sensitive information about individuals, including information about individualsβ identity, health, or demographic information, unless you have obtained the right to do so in accordance with applicable law
5. Engage in or facilitate any action or generate any content that infringes, misappropriates, or otherwise violates any third-party rights, including the outputs or results of any products or services using the Llama Materials
6. Create, generate, or facilitate the creation of malicious code, malware, computer viruses or do anything else that could disable, overburden, interfere with or impair the proper working, integrity, operation or appearance of a website or computer system
7. Engage in any action, or facilitate any action, to intentionally circumvent or remove usage restrictions or other safety measures, or to enable functionality disabled by MetaΒ
2. Engage in, promote, incite, facilitate, or assist in the planning or development of activities that present a risk of death or bodily harm to individuals, including use of Llama 3.2 related to the following:
8. Military, warfare, nuclear industries or applications, espionage, use for materials or activities that are subject to the International Traffic Arms Regulations (ITAR) maintained by the United States Department of State or to the U.S. Biological Weapons Anti-Terrorism Act of 1989 or the Chemical Weapons Convention Implementation Act of 1997
9. Guns and illegal weapons (including weapon development)
10. Illegal drugs and regulated/controlled substances
11. Operation of critical infrastructure, transportation technologies, or heavy machinery
12. Self-harm or harm to others, including suicide, cutting, and eating disorders
13. Any content intended to incite or promote violence, abuse, or any infliction of bodily harm to an individual
3. Intentionally deceive or mislead others, including use of Llama 3.2 related to the following:
14. Generating, promoting, or furthering fraud or the creation or promotion of disinformation
15. Generating, promoting, or furthering defamatory content, including the creation of defamatory statements, images, or other content
16. Generating, promoting, or further distributing spam
17. Impersonating another individual without consent, authorization, or legal right
18. Representing that the use of Llama 3.2 or outputs are human-generated
19. Generating or facilitating false online engagement, including fake reviews and other means of fake online engagementΒ
4. Fail to appropriately disclose to end users any known dangers of your AI system
5. Interact with third party tools, models, or software designed to generate unlawful content or engage in unlawful or harmful conduct and/or represent that the outputs of such tools, models, or software are associated with Meta or Llama 3.2
With respect to any multimodal models included in Llama 3.2, the rights granted under Section 1(a) of the Llama 3.2 Community License Agreement are not being granted to you if you are an individual domiciled in, or a company with a principal place of business in, the European Union. This restriction does not apply to end users of a product or service that incorporates any such multimodal models.
Please report any violation of this Policy, software βbug,β or other problems that could lead to a violation of this Policy through one of the following means:
* Reporting issues with the model: [https://github.com/meta-llama/llama-models/issues](https://l.workplace.com/l.php?u=https%3A%2F%2Fgithub.com%2Fmeta-llama%2Fllama-models%2Fissues&h=AT0qV8W9BFT6NwihiOHRuKYQM_UnkzN_NmHMy91OT55gkLpgi4kQupHUl0ssR4dQsIQ8n3tfd0vtkobvsEvt1l4Ic6GXI2EeuHV8N08OG2WnbAmm0FL4ObkazC6G_256vN0lN9DsykCvCqGZ)
* Reporting risky content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)
* Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)
* Reporting violations of the Acceptable Use Policy or unlicensed uses of Llama 3.2: [email protected]
extra_gated_fields:
First Name: text
Last Name: text
Date of birth: date_picker
Country: country
Affiliation: text
Job title:
type: select
options:
- Student
- Research Graduate
- AI researcher
- AI developer/engineer
- Reporter
- Other
geo: ip_location
By clicking Submit below I accept the terms of the license and acknowledge that the information I provide will be collected stored processed and shared in accordance with the Meta Privacy Policy: checkbox
extra_gated_description: >-
The information you provide will be collected, stored, processed and shared in
accordance with the [Meta Privacy
Policy](https://www.facebook.com/privacy/policy/).
extra_gated_button_content: Submit
---
## Model Information
The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. They outperform many of the available open source and closed chat models on common industry benchmarks.
**Model Developer:** Meta
**Model Architecture:** Llama 3.2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align with human preferences for helpfulness and safety.
| | Training Data | Params | Input modalities | Output modalities | Context Length | GQA | Shared Embeddings | Token count | Knowledge cutoff |
| :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- | :---- |
| Llama 3.2 (text only) | A new mix of publicly available online data. | 1B (1.23B) | Multilingual Text | Multilingual Text and code | 128k | Yes | Yes | Up to 9T tokens | December 2023 |
| | | 3B (3.21B) | Multilingual Text | Multilingual Text and code | | | | | |
| Llama 3.2 Quantized (text only) | A new mix of publicly available online data. | 1B (1.23B) | Multilingual Text | Multilingual Text and code | 8k | Yes | Yes | Up to 9T tokens | December 2023 |
| | | 3B (3.21B) | Multilingual Text | Multilingual Text and code | | | | | |
**Supported Languages:** English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai are officially supported. Llama 3.2 has been trained on a broader collection of languages than these 8 supported languages. Developers may fine-tune Llama 3.2 models for languages beyond these supported languages, provided they comply with the Llama 3.2 Community License and the Acceptable Use Policy. Developers are always expected to ensure that their deployments, including those that involve additional languages, are completed safely and responsibly.
**Llama 3.2 Model Family:** Token counts refer to pretraining data only. All model versions use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Release Date:** Sept 25, 2024
**Status:** This is a static model trained on an offline dataset. Future versions may be released that improve model capabilities and safety.
**License:** Use of Llama 3.2 is governed by the [Llama 3.2 Community License](https://github.com/meta-llama/llama-models/blob/main/models/llama3_2/LICENSE) (a custom, commercial license agreement).
**Feedback:** Instructions on how to provide feedback or comments on the model can be found in the Llama Models [README](https://github.com/meta-llama/llama-models/blob/main/README.md). For more technical information about generation parameters and recipes for how to use Llama 3.2 in applications, please go [here](https://github.com/meta-llama/llama-recipes).
## Intended Use
**Intended Use Cases:** Llama 3.2 is intended for commercial and research use in multiple languages. Instruction tuned text only models are intended for assistant-like chat and agentic applications like knowledge retrieval and summarization, mobile AI powered writing assistants and query and prompt rewriting. Pretrained models can be adapted for a variety of additional natural language generation tasks. Similarly, quantized models can be adapted for a variety of on-device use-cases with limited compute resources.
**Out of Scope:** Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in any other way that is prohibited by the Acceptable Use Policy and Llama 3.2 Community License. Use in languages beyond those explicitly referenced as supported in this model card.
## How to use
This repository contains two versions of Llama-3.2-1B-Instruct, for use with transformers and with the original `llama` codebase.
### Use with transformers
Starting with `transformers >= 4.43.0` onward, you can run conversational inference using the Transformers `pipeline` abstraction or by leveraging the Auto classes with the `generate()` function.
Make sure to update your transformers installation via `pip install --upgrade transformers`.
```python
import torch
from transformers import pipeline
model_id = "meta-llama/Llama-3.2-1B-Instruct"
pipe = pipeline(
"text-generation",
model=model_id,
torch_dtype=torch.bfloat16,
device_map="auto",
)
messages = [
{"role": "system", "content": "You are a pirate chatbot who always responds in pirate speak!"},
{"role": "user", "content": "Who are you?"},
]
outputs = pipe(
messages,
max_new_tokens=256,
)
print(outputs[0]["generated_text"][-1])
```
Note: You can also find detailed recipes on how to use the model locally, with `torch.compile()`, assisted generations, quantised and more at [`huggingface-llama-recipes`](https://github.com/huggingface/huggingface-llama-recipes)
### Use with `llama`
Please, follow the instructions in the [repository](https://github.com/meta-llama/llama)
To download Original checkpoints, see the example command below leveraging `huggingface-cli`:
```
huggingface-cli download meta-llama/Llama-3.2-1B-Instruct --include "original/*" --local-dir Llama-3.2-1B-Instruct
```
## Hardware and Software
**Training Factors:** We used custom training libraries, Meta's custom built GPU cluster, and production infrastructure for pretraining. Fine-tuning, quantization, annotation, and evaluation were also performed on production infrastructure.
**Training Energy Use:** Training utilized a cumulative of **916k** GPU hours of computation on H100-80GB (TDP of 700W) type hardware, per the table below. Training time is the total GPU time required for training each model and power consumption is the peak power capacity per GPU device used, adjusted for power usage efficiency.
**Training Greenhouse Gas Emissions:** Estimated total location-based greenhouse gas emissions were **240** tons CO2eq for training. Since 2020, Meta has maintained net zero greenhouse gas emissions in its global operations and matched 100% of its electricity use with renewable energy; therefore, the total market-based greenhouse gas emissions for training were 0 tons CO2eq.
| | Training Time (GPU hours) | Logit Generation Time (GPU Hours) | Training Power Consumption (W) | Training Location-Based Greenhouse Gas Emissions (tons CO2eq) | Training Market-Based Greenhouse Gas Emissions (tons CO2eq) |
| :---- | :---: | ----- | :---: | :---: | :---: |
| Llama 3.2 1B | 370k | \- | 700 | 107 | 0 |
| Llama 3.2 3B | 460k | \- | 700 | 133 | 0 |
| Llama 3.2 1B SpinQuant | 1.7 | 0 | 700 | *Negligible*\*\* | 0 |
| Llama 3.2 3B SpinQuant | 2.4 | 0 | 700 | *Negligible*\*\* | 0 |
| Llama 3.2 1B QLora | 1.3k | 0 | 700 | 0.381 | 0 |
| Llama 3.2 3B QLora | 1.6k | 0 | 700 | 0.461 | 0 |
| Total | 833k | 86k | | 240 | 0 |
\*\* The location-based CO2e emissions of Llama 3.2 1B SpinQuant and Llama 3.2 3B SpinQuant are less than 0.001 metric tonnes each. This is due to the minimal training GPU hours that are required.
The methodology used to determine training energy use and greenhouse gas emissions can be found [here](https://arxiv.org/pdf/2204.05149). Since Meta is openly releasing these models, the training energy use and greenhouse gas emissions will not be incurred by others.
## Training Data
**Overview:** Llama 3.2 was pretrained on up to 9 trillion tokens of data from publicly available sources. For the 1B and 3B Llama 3.2 models, we incorporated logits from the Llama 3.1 8B and 70B models into the pretraining stage of the model development, where outputs (logits) from these larger models were used as token-level targets. Knowledge distillation was used after pruning to recover performance. In post-training we used a similar recipe as Llama 3.1 and produced final chat models by doing several rounds of alignment on top of the pre-trained model. Each round involved Supervised Fine-Tuning (SFT), Rejection Sampling (RS), and Direct Preference Optimization (DPO).
**Data Freshness:** The pretraining data has a cutoff of December 2023\.
## Quantization
### Quantization Scheme
We designed the current quantization scheme with the [PyTorchβs ExecuTorch](https://github.com/pytorch/executorch) inference framework and Arm CPU backend in mind, taking into account metrics including model quality, prefill/decoding speed, and memory footprint. Our quantization scheme involves three parts:
- All linear layers in all transformer blocks are quantized to a 4-bit groupwise scheme (with a group size of 32) for weights and 8-bit per-token dynamic quantization for activations.
- The classification layer is quantized to 8-bit per-channel for weight and 8-bit per token dynamic quantization for activation.
- Similar to classification layer, an 8-bit per channel quantization is used for embedding layer.
### Quantization-Aware Training and LoRA
The quantization-aware training (QAT) with low-rank adaptation (LoRA) models went through only post-training stages, using the same data as the full precision models. To initialize QAT, we utilize BF16 Llama 3.2 model checkpoints obtained after supervised fine-tuning (SFT) and perform an additional full round of SFT training with QAT. We then freeze the backbone of the QAT model and perform another round of SFT with LoRA adaptors applied to all layers within the transformer block. Meanwhile, the LoRA adaptors' weights and activations are maintained in BF16. Because our approach is similar to QLoRA of Dettmers et al., (2023) (i.e., quantization followed by LoRA adapters), we refer this method as QLoRA. Finally, we fine-tune the resulting model (both backbone and LoRA adaptors) using direct preference optimization (DPO).
### SpinQuant
[SpinQuant](https://arxiv.org/abs/2405.16406) was applied, together with generative post-training quantization (GPTQ). For the SpinQuant rotation matrix fine-tuning, we optimized for 100 iterations, using 800 samples with sequence-length 2048 from the WikiText 2 dataset. For GPTQ, we used 128 samples from the same dataset with the same sequence-length.
## Benchmarks \- English Text
In this section, we report the results for Llama 3.2 models on standard automatic benchmarks. For all these evaluations, we used our internal evaluations library.
### Base Pretrained Models
| Category | Benchmark | \# Shots | Metric | Llama 3.2 1B | Llama 3.2 3B | Llama 3.1 8B |
| ----- | ----- | :---: | :---: | :---: | :---: | :---: |
| General | MMLU | 5 | macro\_avg/acc\_char | 32.2 | 58 | 66.7 |
| | AGIEval English | 3-5 | average/acc\_char | 23.3 | 39.2 | 47.8 |
| | ARC-Challenge | 25 | acc\_char | 32.8 | 69.1 | 79.7 |
| Reading comprehension | SQuAD | 1 | em | 49.2 | 67.7 | 77 |
| | QuAC (F1) | 1 | f1 | 37.9 | 42.9 | 44.9 |
| | DROP (F1) | 3 | f1 | 28.0 | 45.2 | 59.5 |
| Long Context | Needle in Haystack | 0 | em | 96.8 | 1 | 1 |
### Instruction Tuned Models
| Capability | | Benchmark | \# Shots | Metric | Llama 3.2 1B bf16 | Llama 3.2 1B Vanilla PTQ\*\* | Llama 3.2 1B Spin Quant | Llama 3.2 1B QLoRA | Llama 3.2 3B bf16 | Llama 3.2 3B Vanilla PTQ\*\* | Llama 3.2 3B Spin Quant | Llama 3.2 3B QLoRA | Llama 3.1 8B |
| :---: | ----- | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
| General | | MMLU | 5 | macro\_avg/acc | 49.3 | 43.3 | 47.3 | 49.0 | 63.4 | 60.5 | 62 | 62.4 | 69.4 |
| Re-writing | | Open-rewrite eval | 0 | micro\_avg/rougeL | 41.6 | 39.2 | 40.9 | 41.2 | 40.1 | 40.3 | 40.8 | 40.7 | 40.9 |
| Summarization | | TLDR9+ (test) | 1 | rougeL | 16.8 | 14.9 | 16.7 | 16.8 | 19.0 | 19.1 | 19.2 | 19.1 | 17.2 |
| Instruction following | | IFEval | 0 | Avg(Prompt/Instruction acc Loose/Strict) | 59.5 | 51.5 | 58.4 | 55.6 | 77.4 | 73.9 | 73.5 | 75.9 | 80.4 |
| Math | | GSM8K (CoT) | 8 | em\_maj1@1 | 44.4 | 33.1 | 40.6 | 46.5 | 77.7 | 72.9 | 75.7 | 77.9 | 84.5 |
| | | MATH (CoT) | 0 | final\_em | 30.6 | 20.5 | 25.3 | 31.0 | 48.0 | 44.2 | 45.3 | 49.2 | 51.9 |
| Reasoning | | ARC-C | 0 | acc | 59.4 | 54.3 | 57 | 60.7 | 78.6 | 75.6 | 77.6 | 77.6 | 83.4 |
| | | GPQA | 0 | acc | 27.2 | 25.9 | 26.3 | 25.9 | 32.8 | 32.8 | 31.7 | 33.9 | 32.8 |
| | | Hellaswag | 0 | acc | 41.2 | 38.1 | 41.3 | 41.5 | 69.8 | 66.3 | 68 | 66.3 | 78.7 |
| Tool Use | | BFCL V2 | 0 | acc | 25.7 | 14.3 | 15.9 | 23.7 | 67.0 | 53.4 | 60.1 | 63.5 | 67.1 |
| | | Nexus | 0 | macro\_avg/acc | 13.5 | 5.2 | 9.6 | 12.5 | 34.3 | 32.4 | 31.5 | 30.1 | 38.5 |
| Long Context | | InfiniteBench/En.QA | 0 | longbook\_qa/f1 | 20.3 | N/A | N/A | N/A | 19.8 | N/A | N/A | N/A | 27.3 |
| | | InfiniteBench/En.MC | 0 | longbook\_choice/acc | 38.0 | N/A | N/A | N/A | 63.3 | N/A | N/A | N/A | 72.2 |
| | | NIH/Multi-needle | 0 | recall | 75.0 | N/A | N/A | N/A | 84.7 | N/A | N/A | N/A | 98.8 |
| Multilingual | | MGSM (CoT) | 0 | em | 24.5 | 13.7 | 18.2 | 24.4 | 58.2 | 48.9 | 54.3 | 56.8 | 68.9 |
\*\*for comparison purposes only. Model not released.
### Multilingual Benchmarks
| Category | Benchmark | Language | Llama 3.2 1B | Llama 3.2 1B Vanilla PTQ\*\* | Llama 3.2 1B Spin Quant | Llama 3.2 1B QLoRA | Llama 3.2 3B | Llama 3.2 3B Vanilla PTQ\*\* | Llama 3.2 3B Spin Quant | Llama 3.2 3B QLoRA | Llama 3.1 8B |
| :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: | :---: |
| General | MMLU (5-shot, macro_avg/acc) | Portuguese | 39.8 | 34.9 | 38.9 | 40.2 | 54.5 | 50.9 | 53.3 | 53.4 | 62.1 |
| | | Spanish | 41.5 | 36.0 | 39.8 | 41.8 | 55.1 | 51.9 | 53.6 | 53.6 | 62.5 |
| | | Italian | 39.8 | 34.9 | 38.1 | 40.6 | 53.8 | 49.9 | 52.1 | 51.7 | 61.6 |
| | | German | 39.2 | 34.9 | 37.5 | 39.6 | 53.3 | 50.0 | 52.2 | 51.3 | 60.6 |
| | | French | 40.5 | 34.8 | 39.2 | 40.8 | 54.6 | 51.2 | 53.3 | 53.3 | 62.3 |
| | | Hindi | 33.5 | 30.0 | 32.1 | 34.0 | 43.3 | 40.4 | 42.0 | 42.1 | 50.9 |
| | | Thai | 34.7 | 31.2 | 32.4 | 34.9 | 44.5 | 41.3 | 44.0 | 42.2 | 50.3 |
\*\*for comparison purposes only. Model not released.
## Inference time
In the below table, we compare the performance metrics of different quantization methods (SpinQuant and QAT \+ LoRA) with the BF16 baseline. The evaluation was done using the [ExecuTorch](https://github.com/pytorch/executorch) framework as the inference engine, with the ARM CPU as a backend using Android OnePlus 12 device.
| Category | Decode (tokens/sec) | Time-to-first-token (sec) | Prefill (tokens/sec) | Model size (PTE file size in MB) | Memory size (RSS in MB) |
| :---- | ----- | ----- | ----- | ----- | ----- |
| 1B BF16 (baseline) | 19.2 | 1.0 | 60.3 | 2358 | 3,185 |
| 1B SpinQuant | 50.2 (2.6x) | 0.3 (-76.9%) | 260.5 (4.3x) | 1083 (-54.1%) | 1,921 (-39.7%) |
| 1B QLoRA | 45.8 (2.4x) | 0.3 (-76.0%) | 252.0 (4.2x) | 1127 (-52.2%) | 2,255 (-29.2%) |
| 3B BF16 (baseline) | 7.6 | 3.0 | 21.2 | 6129 | 7,419 |
| 3B SpinQuant | 19.7 (2.6x) | 0.7 (-76.4%) | 89.7 (4.2x) | 2435 (-60.3%) | 3,726 (-49.8%) |
| 3B QLoRA | 18.5 (2.4x) | 0.7 (-76.1%) | 88.8 (4.2x) | 2529 (-58.7%) | 4,060 (-45.3%) |
(\*) The performance measurement is done using an adb binary-based approach.
(\*\*) It is measured on an Android OnePlus 12 device.
(\*\*\*) Time-to-first-token (TTFT) is measured with prompt length=64
*Footnote:*
- *Decode (tokens/second) is for how quickly it keeps generating. Higher is better.*
- *Time-to-first-token (TTFT for shorthand) is for how fast it generates the first token for a given prompt. Lower is better.*
- *Prefill is the inverse of TTFT (aka 1/TTFT) in tokens/second. Higher is better*
- *Model size \- how big is the model, measured by, PTE file, a binary file format for ExecuTorch*
- *RSS size \- Memory usage in resident set size (RSS)*
## Responsibility & Safety
As part of our Responsible release approach, we followed a three-pronged strategy to managing trust & safety risks:
1. Enable developers to deploy helpful, safe and flexible experiences for their target audience and for the use cases supported by Llama
2. Protect developers against adversarial users aiming to exploit Llama capabilities to potentially cause harm
3. Provide protections for the community to help prevent the misuse of our models
### Responsible Deployment
**Approach:** Llama is a foundational technology designed to be used in a variety of use cases. Examples on how Metaβs Llama models have been responsibly deployed can be found in our [Community Stories webpage](https://llama.meta.com/community-stories/). Our approach is to build the most helpful models, enabling the world to benefit from the technology power, by aligning our model safety for generic use cases and addressing a standard set of harms. Developers are then in the driverβs seat to tailor safety for their use cases, defining their own policies and deploying the models with the necessary safeguards in their Llama systems. Llama 3.2 was developed following the best practices outlined in our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide/).
#### Llama 3.2 Instruct
**Objective:** Our main objectives for conducting safety fine-tuning are to provide the research community with a valuable resource for studying the robustness of safety fine-tuning, as well as to offer developers a readily available, safe, and powerful model for various applications to reduce the developer workload to deploy safe AI systems. We implemented the same set of safety mitigations as in Llama 3, and you can learn more about these in the Llama 3 [paper](https://ai.meta.com/research/publications/the-llama-3-herd-of-models/).
**Fine-Tuning Data:** We employ a multi-faceted approach to data collection, combining human-generated data from our vendors with synthetic data to mitigate potential safety risks. Weβve developed many large language model (LLM)-based classifiers that enable us to thoughtfully select high-quality prompts and responses, enhancing data quality control.
**Refusals and Tone:** Building on the work we started with Llama 3, we put a great emphasis on model refusals to benign prompts as well as refusal tone. We included both borderline and adversarial prompts in our safety data strategy, and modified our safety data responses to follow tone guidelines.
#### Llama 3.2 Systems
**Safety as a System:** Large language models, including Llama 3.2, **are not designed to be deployed in isolation** but instead should be deployed as part of an overall AI system with additional safety guardrails as required. Developers are expected to deploy system safeguards when building agentic systems. Safeguards are key to achieve the right helpfulness-safety alignment as well as mitigating safety and security risks inherent to the system and any integration of the model or system with external tools. As part of our responsible release approach, we provide the community with [safeguards](https://llama.meta.com/trust-and-safety/) that developers should deploy with Llama models or other LLMs, including Llama Guard, Prompt Guard and Code Shield. All our [reference implementations](https://github.com/meta-llama/llama-agentic-system) demos contain these safeguards by default so developers can benefit from system-level safety out-of-the-box.
### New Capabilities and Use Cases
**Technological Advancement:** Llama releases usually introduce new capabilities that require specific considerations in addition to the best practices that generally apply across all Generative AI use cases. For prior release capabilities also supported by Llama 3.2, see [Llama 3.1 Model Card](https://github.com/meta-llama/llama-models/blob/main/models/llama3_1/MODEL_CARD.md), as the same considerations apply here as well.
**Constrained Environments:** Llama 3.2 1B and 3B models are expected to be deployed in highly constrained environments, such as mobile devices. LLM Systems using smaller models will have a different alignment profile and safety/helpfulness tradeoff than more complex, larger systems. Developers should ensure the safety of their system meets the requirements of their use case. We recommend using lighter system safeguards for such use cases, like Llama Guard 3-1B or its mobile-optimized version.
### Evaluations
**Scaled Evaluations:** We built dedicated, adversarial evaluation datasets and evaluated systems composed of Llama models and Purple Llama safeguards to filter input prompt and output response. It is important to evaluate applications in context, and we recommend building dedicated evaluation dataset for your use case.
**Red Teaming:** We conducted recurring red teaming exercises with the goal of discovering risks via adversarial prompting and we used the learnings to improve our benchmarks and safety tuning datasets. We partnered early with subject-matter experts in critical risk areas to understand the nature of these real-world harms and how such models may lead to unintended harm for society. Based on these conversations, we derived a set of adversarial goals for the red team to attempt to achieve, such as extracting harmful information or reprogramming the model to act in a potentially harmful capacity. The red team consisted of experts in cybersecurity, adversarial machine learning, responsible AI, and integrity in addition to multilingual content specialists with background in integrity issues in specific geographic markets.
### Critical Risks
In addition to our safety work above, we took extra care on measuring and/or mitigating the following critical risk areas:
**1\. CBRNE (Chemical, Biological, Radiological, Nuclear, and Explosive Weapons):** Llama 3.2 1B and 3B models are smaller and less capable derivatives of Llama 3.1. For Llama 3.1 70B and 405B, to assess risks related to proliferation of chemical and biological weapons, we performed uplift testing designed to assess whether use of Llama 3.1 models could meaningfully increase the capabilities of malicious actors to plan or carry out attacks using these types of weapons and have determined that such testing also applies to the smaller 1B and 3B models.
**2\. Child Safety:** Child Safety risk assessments were conducted using a team of experts, to assess the modelβs capability to produce outputs that could result in Child Safety risks and inform on any necessary and appropriate risk mitigations via fine tuning. We leveraged those expert red teaming sessions to expand the coverage of our evaluation benchmarks through Llama 3 model development. For Llama 3, we conducted new in-depth sessions using objective based methodologies to assess the model risks along multiple attack vectors including the additional languages Llama 3 is trained on. We also partnered with content specialists to perform red teaming exercises assessing potentially violating content while taking account of market specific nuances or experiences.
**3\. Cyber Attacks:** For Llama 3.1 405B, our cyber attack uplift study investigated whether LLMs can enhance human capabilities in hacking tasks, both in terms of skill level and speed.
Our attack automation study focused on evaluating the capabilities of LLMs when used as autonomous agents in cyber offensive operations, specifically in the context of ransomware attacks. This evaluation was distinct from previous studies that considered LLMs as interactive assistants. The primary objective was to assess whether these models could effectively function as independent agents in executing complex cyber-attacks without human intervention. Because Llama 3.2βs 1B and 3B models are smaller and less capable models than Llama 3.1 405B, we broadly believe that the testing conducted for the 405B model also applies to Llama 3.2 models.
### Community
**Industry Partnerships:** Generative AI safety requires expertise and tooling, and we believe in the strength of the open community to accelerate its progress. We are active members of open consortiums, including the AI Alliance, Partnership on AI and MLCommons, actively contributing to safety standardization and transparency. We encourage the community to adopt taxonomies like the MLCommons Proof of Concept evaluation to facilitate collaboration and transparency on safety and content evaluations. Our Purple Llama tools are open sourced for the community to use and widely distributed across ecosystem partners including cloud service providers. We encourage community contributions to our [Github repository](https://github.com/meta-llama/PurpleLlama).
**Grants:** We also set up the [Llama Impact Grants](https://llama.meta.com/llama-impact-grants/) program to identify and support the most compelling applications of Metaβs Llama model for societal benefit across three categories: education, climate and open innovation. The 20 finalists from the hundreds of applications can be found [here](https://llama.meta.com/llama-impact-grants/#finalists).
**Reporting:** Finally, we put in place a set of resources including an [output reporting mechanism](https://developers.facebook.com/llama_output_feedback) and [bug bounty program](https://www.facebook.com/whitehat) to continuously improve the Llama technology with the help of the community.
## Ethical Considerations and Limitations
**Values:** The core values of Llama 3.2 are openness, inclusivity and helpfulness. It is meant to serve everyone, and to work for a wide range of use cases. It is thus designed to be accessible to people across many different backgrounds, experiences and perspectives. Llama 3.2 addresses users and their needs as they are, without insertion unnecessary judgment or normativity, while reflecting the understanding that even content that may appear problematic in some cases can serve valuable purposes in others. It respects the dignity and autonomy of all users, especially in terms of the values of free thought and expression that power innovation and progress.
**Testing:** Llama 3.2 is a new technology, and like any new technology, there are risks associated with its use. Testing conducted to date has not covered, nor could it cover, all scenarios. For these reasons, as with all LLMs, Llama 3.2βs potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 3.2 models, developers should perform safety testing and tuning tailored to their specific applications of the model. Please refer to available resources including our [Responsible Use Guide](https://llama.meta.com/responsible-use-guide), [Trust and Safety](https://llama.meta.com/trust-and-safety/) solutions, and other [resources](https://llama.meta.com/docs/get-started/) to learn more about responsible development.
|
RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf
|
RichardErkhov
| 2025-04-28T22:05:40Z | 0 | 0 | null |
[
"gguf",
"endpoints_compatible",
"region:us",
"conversational"
] | null | 2025-04-28T20:29:31Z |
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B - GGUF
- Model creator: https://huggingface.co/cutelemonlili/
- Original model: https://huggingface.co/cutelemonlili/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q2_K.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q2_K.gguf) | Q2_K | 2.81GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ3_XS.gguf) | IQ3_XS | 3.12GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ3_S.gguf) | IQ3_S | 3.26GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K_S.gguf) | Q3_K_S | 3.25GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ3_M.gguf) | IQ3_M | 3.33GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K.gguf) | Q3_K | 3.55GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K_M.gguf) | Q3_K_M | 3.55GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q3_K_L.gguf) | Q3_K_L | 3.81GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ4_XS.gguf) | IQ4_XS | 3.96GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_0.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_0.gguf) | Q4_0 | 4.13GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.IQ4_NL.gguf) | IQ4_NL | 4.16GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_K_S.gguf) | Q4_K_S | 4.15GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_K.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_K.gguf) | Q4_K | 4.36GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_K_M.gguf) | Q4_K_M | 4.36GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_1.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q4_1.gguf) | Q4_1 | 4.54GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_0.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_0.gguf) | Q5_0 | 4.95GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_K_S.gguf) | Q5_K_S | 4.95GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_K.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_K.gguf) | Q5_K | 5.07GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_K_M.gguf) | Q5_K_M | 5.07GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_1.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q5_1.gguf) | Q5_1 | 5.36GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q6_K.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q6_K.gguf) | Q6_K | 5.82GB |
| [Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q8_0.gguf](https://huggingface.co/RichardErkhov/cutelemonlili_-_Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B-gguf/blob/main/Qwen2.5-7B-Instruct_MATH_training_response_Qwen2.5_7B.Q8_0.gguf) | Q8_0 | 7.54GB |
Original model description:
---
library_name: transformers
license: other
base_model: Qwen/Qwen2.5-7B-Instruct
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: MATH_training_response_Qwen2.5_7B
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# MATH_training_response_Qwen2.5_7B
This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the MATH_training_response_Qwen2.5_7B dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0941
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 8
- total_eval_batch_size: 4
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 0.1184 | 0.3559 | 200 | 0.1123 |
| 0.1292 | 0.7117 | 400 | 0.1061 |
| 0.0325 | 1.0676 | 600 | 0.1058 |
| 0.0278 | 1.4235 | 800 | 0.1001 |
| 0.0421 | 1.7794 | 1000 | 0.0947 |
### Framework versions
- Transformers 4.46.1
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.20.3
|
taybet/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bellowing_sedate_otter
|
taybet
| 2025-04-28T22:01:11Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"generated_from_trainer",
"rl-swarm",
"grpo",
"gensyn",
"I am bellowing sedate otter",
"trl",
"conversational",
"arxiv:2402.03300",
"base_model:Gensyn/Qwen2.5-0.5B-Instruct",
"base_model:finetune:Gensyn/Qwen2.5-0.5B-Instruct",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-24T07:26:37Z |
---
base_model: Gensyn/Qwen2.5-0.5B-Instruct
library_name: transformers
model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bellowing_sedate_otter
tags:
- generated_from_trainer
- rl-swarm
- grpo
- gensyn
- I am bellowing sedate otter
- trl
licence: license
---
# Model Card for Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bellowing_sedate_otter
This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="taybet/Qwen2.5-0.5B-Instruct-Gensyn-Swarm-bellowing_sedate_otter", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.7.0
- Datasets: 3.5.0
- Tokenizers: 0.21.1
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouΓ©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
JohnConnor123/Qwen2.5-0.5B-Instruct-BNB-8bit
|
JohnConnor123
| 2025-04-28T21:58:34Z | 6 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"chat",
"conversational",
"en",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-0.5B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-0.5B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"8-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2025-04-28T01:26:02Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE
language:
- en
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-0.5B-Instruct
tags:
- chat
library_name: transformers
---
> ## **This quantization was done using the [quantization-benchmark](https://github.com/JohnConnor123/quantization-benchmark) framework**
## Bitsandbytes quantization config
>{'load_in_8bit': True}
# Qwen2.5-0.5B-Instruct
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the instruction-tuned 0.5B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
- Number of Parameters: 0.49B
- Number of Paramaters (Non-Embedding): 0.36B
- Number of Layers: 24
- Number of Attention Heads (GQA): 14 for Q and 2 for KV
- Context Length: Full 32,768 tokens and generation 8192 tokens
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen2.5-0.5B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
## Evaluation & Performance
Detailed evaluation results are reported in this [π blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
JohnConnor123/Qwen2.5-0.5B-Instruct-BNB-4bit
|
JohnConnor123
| 2025-04-28T21:58:20Z | 9 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"chat",
"conversational",
"en",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-0.5B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-0.5B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"bitsandbytes",
"region:us"
] |
text-generation
| 2025-01-09T11:21:11Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE
language:
- en
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-0.5B-Instruct
tags:
- chat
library_name: transformers
---
> ## **This quantization was done using the [quantization-benchmark](https://github.com/JohnConnor123/quantization-benchmark) framework**
## Bitsandbytes quantization config
>{'load_in_4bit': True}
# Qwen2.5-0.5B-Instruct
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the instruction-tuned 0.5B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
- Number of Parameters: 0.49B
- Number of Paramaters (Non-Embedding): 0.36B
- Number of Layers: 24
- Number of Attention Heads (GQA): 14 for Q and 2 for KV
- Context Length: Full 32,768 tokens and generation 8192 tokens
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen2.5-0.5B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
## Evaluation & Performance
Detailed evaluation results are reported in this [π blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
JohnConnor123/Qwen2.5-0.5B-Instruct-AWQ-64G-INT4-vGEMM
|
JohnConnor123
| 2025-04-28T21:58:04Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen2",
"text-generation",
"chat",
"conversational",
"en",
"arxiv:2407.10671",
"base_model:Qwen/Qwen2.5-0.5B-Instruct",
"base_model:quantized:Qwen/Qwen2.5-0.5B-Instruct",
"license:apache-2.0",
"autotrain_compatible",
"text-generation-inference",
"endpoints_compatible",
"4-bit",
"awq",
"region:us"
] |
text-generation
| 2025-04-28T01:25:05Z |
---
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE
language:
- en
pipeline_tag: text-generation
base_model: Qwen/Qwen2.5-0.5B-Instruct
tags:
- chat
library_name: transformers
---
> ## **This quantization was done using the [quantization-benchmark](https://github.com/JohnConnor123/quantization-benchmark) framework**
## AWQ quantization config
>{'w_bit': 4, 'q_group_size': 64, 'zero_point': True, 'version': 'GEMM'}
# Qwen2.5-0.5B-Instruct
## Introduction
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
- Significantly **more knowledge** and has greatly improved capabilities in **coding** and **mathematics**, thanks to our specialized expert models in these domains.
- Significant improvements in **instruction following**, **generating long texts** (over 8K tokens), **understanding structured data** (e.g, tables), and **generating structured outputs** especially JSON. **More resilient to the diversity of system prompts**, enhancing role-play implementation and condition-setting for chatbots.
- **Long-context Support** up to 128K tokens and can generate up to 8K tokens.
- **Multilingual support** for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.
**This repo contains the instruction-tuned 0.5B Qwen2.5 model**, which has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
- Number of Parameters: 0.49B
- Number of Paramaters (Non-Embedding): 0.36B
- Number of Layers: 24
- Number of Attention Heads (GQA): 14 for Q and 2 for KV
- Context Length: Full 32,768 tokens and generation 8192 tokens
For more details, please refer to our [blog](https://qwenlm.github.io/blog/qwen2.5/), [GitHub](https://github.com/QwenLM/Qwen2.5), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Requirements
The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.37.0`, you will encounter the following error:
```
KeyError: 'qwen2'
```
## Quickstart
Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen2.5-0.5B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=512
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
```
## Evaluation & Performance
Detailed evaluation results are reported in this [π blog](https://qwenlm.github.io/blog/qwen2.5/).
For requirements on GPU memory and the respective throughput, see results [here](https://qwen.readthedocs.io/en/latest/benchmark/speed_benchmark.html).
## Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen2.5,
title = {Qwen2.5: A Party of Foundation Models},
url = {https://qwenlm.github.io/blog/qwen2.5/},
author = {Qwen Team},
month = {September},
year = {2024}
}
@article{qwen2,
title={Qwen2 Technical Report},
author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
journal={arXiv preprint arXiv:2407.10671},
year={2024}
}
```
|
fffanx/Llama-3.2-1B-Instruct-GRPO-agent18
|
fffanx
| 2025-04-28T21:55:32Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"grpo",
"dataset:groupd_dataset",
"arxiv:2402.03300",
"base_model:meta-llama/Llama-3.2-1B-Instruct",
"base_model:finetune:meta-llama/Llama-3.2-1B-Instruct",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T21:44:27Z |
---
base_model: meta-llama/Llama-3.2-1B-Instruct
datasets: groupd_dataset
library_name: transformers
model_name: Llama-3.2-1B-Instruct-GRPO-agent18
tags:
- generated_from_trainer
- trl
- grpo
licence: license
---
# Model Card for Llama-3.2-1B-Instruct-GRPO-agent18
This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on the [groupd_dataset](https://huggingface.co/datasets/groupd_dataset) dataset.
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="fffanx/Llama-3.2-1B-Instruct-GRPO-agent18", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
### Framework versions
- TRL: 0.17.0.dev0
- Transformers: 4.49.0
- Pytorch: 2.6.0
- Datasets: 3.3.2
- Tokenizers: 0.21.0
## Citations
Cite GRPO as:
```bibtex
@article{zhihong2024deepseekmath,
title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
year = 2024,
eprint = {arXiv:2402.03300},
}
```
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
IHe-KaiI/CTRL-D
|
IHe-KaiI
| 2025-04-28T21:54:08Z | 0 | 0 |
diffusers
|
[
"diffusers",
"safetensors",
"license:apache-2.0",
"region:us"
] | null | 2025-03-06T04:38:05Z |
---
license: apache-2.0
---
|
imasaki/zhongligemma
|
imasaki
| 2025-04-28T21:54:05Z | 0 | 0 |
transformers
|
[
"transformers",
"tensorboard",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:google/gemma-3-27b-pt",
"base_model:finetune:google/gemma-3-27b-pt",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T08:49:34Z |
---
base_model: google/gemma-3-27b-pt
library_name: transformers
model_name: zhongligemma
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for zhongligemma
This model is a fine-tuned version of [google/gemma-3-27b-pt](https://huggingface.co/google/gemma-3-27b-pt).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="imasaki/zhongligemma", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
This model was trained with SFT.
### Framework versions
- TRL: 0.15.2
- Transformers: 4.51.3
- Pytorch: 2.7.0+cu128
- Datasets: 3.3.2
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin GallouοΎο½©dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
JohnConnor123/Qwen3-0.6B-Q8_0
|
JohnConnor123
| 2025-04-28T21:51:34Z | 0 | 0 |
transformers
|
[
"transformers",
"gguf",
"qwen3",
"text-generation",
"conversational",
"base_model:Qwen/Qwen3-0.6B",
"base_model:quantized:Qwen/Qwen3-0.6B",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T21:24:12Z |
---
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-0.6B/blob/main/LICENSE
pipeline_tag: text-generation
base_model: Qwen/Qwen3-0.6B
---
> ## **This quantization was done using the [quantization-benchmark](https://github.com/JohnConnor123/quantization-benchmark) framework**
# Qwen3-0.6B
<a href="https://chat.qwen.ai/" target="_blank" style="margin: 2px;">
<img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
</a>
## Qwen3 Highlights
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
- **Uniquely support of seamless switching between thinking mode** (for complex logical reasoning, math, and coding) and **non-thinking mode** (for efficient, general-purpose dialogue) **within single model**, ensuring optimal performance across various scenarios.
- **Significantly enhancement in its reasoning capabilities**, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
- **Superior human preference alignment**, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
- **Expertise in agent capabilities**, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
- **Support of 100+ languages and dialects** with strong capabilities for **multilingual instruction following** and **translation**.
## Model Overview
**Qwen3-0.6B** has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Number of Parameters: 0.6B
- Number of Paramaters (Non-Embedding): 0.44B
- Number of Layers: 28
- Number of Attention Heads (GQA): 16 for Q and 8 for KV
- Context Length: 32,768
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [blog](https://qwenlm.github.io/blog/qwen3/), [GitHub](https://github.com/QwenLM/Qwen3), and [Documentation](https://qwen.readthedocs.io/en/latest/).
> [!TIP]
> If you encounter significant endless repetitions, please refer to the [Best Practices](#best-practices) section for optimal sampling parameters, and set the ``presence_penalty`` to 1.5.
## Quickstart
The code of Qwen3 has been in the latest Hugging Face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.51.0`, you will encounter the following error:
```
KeyError: 'qwen3'
```
The following contains a code snippet illustrating how to use the model generate content based on given inputs.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen3-0.6B"
# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
# conduct text completion
generated_ids = model.generate(
**model_inputs,
max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
# parsing thinking content
try:
# rindex finding 151668 (</think>)
index = len(output_ids) - output_ids[::-1].index(151668)
except ValueError:
index = 0
thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
print("thinking content:", thinking_content)
print("content:", content)
```
For deployment, you can use `sglang>=0.4.6.post1` or `vllm>=0.8.4` or to create an OpenAI-compatible API endpoint:
- SGLang:
```shell
python -m sglang.launch_server --model-path Qwen/Qwen3-0.6B --reasoning-parser qwen3
```
- vLLM:
```shell
vllm serve Qwen/Qwen3-0.6B --enable-reasoning --reasoning-parser deepseek_r1
```
For local use, applications such as llama.cpp, Ollama, LMStudio, and MLX-LM have also supported Qwen3.
## Switching Between Thinking and Non-Thinking Mode
> [!TIP]
> The `enable_thinking` switch is also available in APIs created by SGLang and vLLM.
> Please refer to our documentation for [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) and [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) users.
### `enable_thinking=True`
By default, Qwen3 has thinking capabilities enabled, similar to QwQ-32B. This means the model will use its reasoning abilities to enhance the quality of generated responses. For example, when explicitly setting `enable_thinking=True` or leaving it as the default value in `tokenizer.apply_chat_template`, the model will engage its thinking mode.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # True is the default value for enable_thinking
)
```
In this mode, the model will generate think content wrapped in a `<think>...</think>` block, followed by the final response.
> [!NOTE]
> For thinking mode, use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0` (the default setting in `generation_config.json`). **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### `enable_thinking=False`
We provide a hard switch to strictly disable the model's thinking behavior, aligning its functionality with the previous Qwen2.5-Instruct models. This mode is particularly useful in scenarios where disabling thinking is essential for enhancing efficiency.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=False # Setting enable_thinking=False disables thinking mode
)
```
In this mode, the model will not generate any think content and will not include a `<think>...</think>` block.
> [!NOTE]
> For non-thinking mode, we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### Advanced Usage: Switching Between Thinking and Non-Thinking Modes via User Input
We provide a soft switch mechanism that allows users to dynamically control the model's behavior when `enable_thinking=True`. Specifically, you can add `/think` and `/no_think` to user prompts or system messages to switch the model's thinking mode from turn to turn. The model will follow the most recent instruction in multi-turn conversations.
Here is an example of a multi-turn conversation:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
class QwenChatbot:
def __init__(self, model_name="Qwen/Qwen3-0.6B"):
self.tokenizer = AutoTokenizer.from_pretrained(model_name)
self.model = AutoModelForCausalLM.from_pretrained(model_name)
self.history = []
def generate_response(self, user_input):
messages = self.history + [{"role": "user", "content": user_input}]
text = self.tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
inputs = self.tokenizer(text, return_tensors="pt")
response_ids = self.model.generate(**inputs, max_new_tokens=32768)[0][len(inputs.input_ids[0]):].tolist()
response = self.tokenizer.decode(response_ids, skip_special_tokens=True)
# Update history
self.history.append({"role": "user", "content": user_input})
self.history.append({"role": "assistant", "content": response})
return response
# Example Usage
if __name__ == "__main__":
chatbot = QwenChatbot()
# First input (without /think or /no_think tags, thinking mode is enabled by default)
user_input_1 = "How many r's in strawberries?"
print(f"User: {user_input_1}")
response_1 = chatbot.generate_response(user_input_1)
print(f"Bot: {response_1}")
print("----------------------")
# Second input with /no_think
user_input_2 = "Then, how many r's in blueberries? /no_think"
print(f"User: {user_input_2}")
response_2 = chatbot.generate_response(user_input_2)
print(f"Bot: {response_2}")
print("----------------------")
# Third input with /think
user_input_3 = "Really? /think"
print(f"User: {user_input_3}")
response_3 = chatbot.generate_response(user_input_3)
print(f"Bot: {response_3}")
```
> [!NOTE]
> For API compatibility, when `enable_thinking=True`, regardless of whether the user uses `/think` or `/no_think`, the model will always output a block wrapped in `<think>...</think>`. However, the content inside this block may be empty if thinking is disabled.
> When `enable_thinking=False`, the soft switches are not valid. Regardless of any `/think` or `/no_think` tags input by the user, the model will not generate think content and will not include a `<think>...</think>` block.
## Agentic Use
Qwen3 excels in tool calling capabilities. We recommend using [Qwen-Agent](https://github.com/QwenLM/Qwen-Agent) to make the best use of agentic ability of Qwen3. Qwen-Agent encapsulates tool-calling templates and tool-calling parsers internally, greatly reducing coding complexity.
To define the available tools, you can use the MCP configuration file, use the integrated tool of Qwen-Agent, or integrate other tools by yourself.
```python
from qwen_agent.agents import Assistant
# Define LLM
llm_cfg = {
'model': 'Qwen3-0.6B',
# Use the endpoint provided by Alibaba Model Studio:
# 'model_type': 'qwen_dashscope',
# 'api_key': os.getenv('DASHSCOPE_API_KEY'),
# Use a custom endpoint compatible with OpenAI API:
'model_server': 'http://localhost:8000/v1', # api_base
'api_key': 'EMPTY',
# Other parameters:
# 'generate_cfg': {
# # Add: When the response content is `<think>this is the thought</think>this is the answer;
# # Do not add: When the response has been separated by reasoning_content and content.
# 'thought_in_content': True,
# },
}
# Define Tools
tools = [
{'mcpServers': { # You can specify the MCP configuration file
'time': {
'command': 'uvx',
'args': ['mcp-server-time', '--local-timezone=Asia/Shanghai']
},
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
},
'code_interpreter', # Built-in tools
]
# Define Agent
bot = Assistant(llm=llm_cfg, function_list=tools)
# Streaming generation
messages = [{'role': 'user', 'content': 'https://qwenlm.github.io/blog/ Introduce the latest developments of Qwen'}]
for responses in bot.run(messages=messages):
pass
print(responses)
```
## Best Practices
To achieve optimal performance, we recommend the following settings:
1. **Sampling Parameters**:
- For thinking mode (`enable_thinking=True`), use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0`. **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions.
- For non-thinking mode (`enable_thinking=False`), we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`.
- For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
2. **Adequate Output Length**: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 38,912 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
3. **Standardize Output Format**: We recommend using prompts to standardize model outputs when benchmarking.
- **Math Problems**: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
- **Multiple-Choice Questions**: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the `answer` field with only the choice letter, e.g., `"answer": "C"`."
4. **No Thinking Content in History**: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
### Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen3,
title = {Qwen3},
url = {https://qwenlm.github.io/blog/qwen3/},
author = {Qwen Team},
month = {April},
year = {2025}
}
```
|
reecursion/llama-3.1-8b-cb50-scm-dualscaffolding
|
reecursion
| 2025-04-28T21:51:23Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"generated_from_trainer",
"trl",
"sft",
"base_model:meta-llama/Llama-3.1-8B",
"base_model:finetune:meta-llama/Llama-3.1-8B",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T20:16:40Z |
---
base_model: meta-llama/Meta-Llama-3.1-8B
library_name: transformers
model_name: llama-3.1-8b-cb50-scm-dualscaffolding
tags:
- generated_from_trainer
- trl
- sft
licence: license
---
# Model Card for llama-3.1-8b-cb50-scm-dualscaffolding
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B).
It has been trained using [TRL](https://github.com/huggingface/trl).
## Quick start
```python
from transformers import pipeline
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="reecursion/llama-3.1-8b-cb50-scm-dualscaffolding", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```
## Training procedure
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/gganeshl-carnegie-mellon-university/Llama-Finetuning/runs/ujj3k6th)
This model was trained with SFT.
### Framework versions
- TRL: 0.17.0
- Transformers: 4.51.3
- Pytorch: 2.6.0
- Datasets: 3.5.1
- Tokenizers: 0.21.1
## Citations
Cite TRL as:
```bibtex
@misc{vonwerra2022trl,
title = {{TRL: Transformer Reinforcement Learning}},
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
year = 2020,
journal = {GitHub repository},
publisher = {GitHub},
howpublished = {\url{https://github.com/huggingface/trl}}
}
```
|
mlx-community/Qwen3-8B-4bit
|
mlx-community
| 2025-04-28T21:51:20Z | 0 | 0 |
mlx
|
[
"mlx",
"safetensors",
"qwen3",
"text-generation",
"conversational",
"base_model:Qwen/Qwen3-8B",
"base_model:quantized:Qwen/Qwen3-8B",
"license:apache-2.0",
"4-bit",
"region:us"
] |
text-generation
| 2025-04-28T21:44:32Z |
---
library_name: mlx
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-8B/blob/main/LICENSE
pipeline_tag: text-generation
base_model: Qwen/Qwen3-8B
tags:
- mlx
---
# mlx-community/Qwen3-8B-4bit
This model [mlx-community/Qwen3-8B-4bit](https://huggingface.co/mlx-community/Qwen3-8B-4bit) was
converted to MLX format from [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B)
using mlx-lm version **0.24.0**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Qwen3-8B-4bit")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
```
|
mchl914/Llama-3.1-Panacea-8B-Instruct
|
mchl914
| 2025-04-28T21:51:12Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"text-generation-inference",
"unsloth",
"llama",
"trl",
"en",
"base_model:wzqacky/Llama-3.1-Panacea-8B",
"base_model:finetune:wzqacky/Llama-3.1-Panacea-8B",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | null | 2025-04-28T21:18:37Z |
---
base_model: wzqacky/Llama-3.1-Panacea-8B
tags:
- text-generation-inference
- transformers
- unsloth
- llama
- trl
license: apache-2.0
language:
- en
---
# Uploaded model
- **Developed by:** mchl914
- **License:** apache-2.0
- **Finetuned from model :** wzqacky/Llama-3.1-Panacea-8B
This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
pedalnomica/Qwen3-235B-A22B
|
pedalnomica
| 2025-04-28T21:44:04Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen3_moe",
"text-generation",
"conversational",
"arxiv:2309.00071",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T21:43:52Z |
---
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-235B-A22B/blob/main/LICENSE
pipeline_tag: text-generation
---
# Qwen3-235B-A22B
<a href="https://chat.qwen.ai/" target="_blank" style="margin: 2px;">
<img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
</a>
## Qwen3 Highlights
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
- **Uniquely support of seamless switching between thinking mode** (for complex logical reasoning, math, and coding) and **non-thinking mode** (for efficient, general-purpose dialogue) **within single model**, ensuring optimal performance across various scenarios.
- **Significantly enhancement in its reasoning capabilities**, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
- **Superior human preference alignment**, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
- **Expertise in agent capabilities**, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
- **Support of 100+ languages and dialects** with strong capabilities for **multilingual instruction following** and **translation**.
## Model Overview
**Qwen3-235B-A22B** has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Number of Parameters: 235B in total and 22B activated
- Number of Paramaters (Non-Embedding): 234B
- Number of Layers: 94
- Number of Attention Heads (GQA): 64 for Q and 4 for KV
- Number of Experts: 128
- Number of Activated Experts: 8
- Context Length: 32,768 natively and [131,072 tokens with YaRN](#processing-long-texts).
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [blog](https://qwenlm.github.io/blog/qwen3/), [GitHub](https://github.com/QwenLM/Qwen3), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Quickstart
The code of Qwen3-MoE has been in the latest Hugging Face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.51.0`, you will encounter the following error:
```
KeyError: 'qwen3_moe'
```
The following contains a code snippet illustrating how to use the model generate content based on given inputs.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen3-235B-A22B"
# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
# conduct text completion
generated_ids = model.generate(
**model_inputs,
max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
# parsing thinking content
try:
# rindex finding 151668 (</think>)
index = len(output_ids) - output_ids[::-1].index(151668)
except ValueError:
index = 0
thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
print("thinking content:", thinking_content)
print("content:", content)
```
For deployment, you can use `sglang>=0.4.6.post1` or `vllm>=0.8.4` or to create an OpenAI-compatible API endpoint:
- SGLang:
```shell
python -m sglang.launch_server --model-path Qwen/Qwen3-235B-A22B --reasoning-parser qwen3
```
- vLLM:
```shell
vllm serve Qwen/Qwen3-235B-A22B --enable-reasoning --reasoning-parser deepseek_r1
```
For local use, applications such as llama.cpp, Ollama, LMStudio, and MLX-LM have also supported Qwen3.
## Switching Between Thinking and Non-Thinking Mode
> [!TIP]
> The `enable_thinking` switch is also available in APIs created by SGLang and vLLM.
> Please refer to our documentation for [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) and [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) users.
### `enable_thinking=True`
By default, Qwen3 has thinking capabilities enabled, similar to QwQ-32B. This means the model will use its reasoning abilities to enhance the quality of generated responses. For example, when explicitly setting `enable_thinking=True` or leaving it as the default value in `tokenizer.apply_chat_template`, the model will engage its thinking mode.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # True is the default value for enable_thinking
)
```
In this mode, the model will generate think content wrapped in a `<think>...</think>` block, followed by the final response.
> [!NOTE]
> For thinking mode, use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0` (the default setting in `generation_config.json`). **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### `enable_thinking=False`
We provide a hard switch to strictly disable the model's thinking behavior, aligning its functionality with the previous Qwen2.5-Instruct models. This mode is particularly useful in scenarios where disabling thinking is essential for enhancing efficiency.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=False # Setting enable_thinking=False disables thinking mode
)
```
In this mode, the model will not generate any think content and will not include a `<think>...</think>` block.
> [!NOTE]
> For non-thinking mode, we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### Advanced Usage: Switching Between Thinking and Non-Thinking Modes via User Input
We provide a soft switch mechanism that allows users to dynamically control the model's behavior when `enable_thinking=True`. Specifically, you can add `/think` and `/no_think` to user prompts or system messages to switch the model's thinking mode from turn to turn. The model will follow the most recent instruction in multi-turn conversations.
Here is an example of a multi-turn conversation:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
class QwenChatbot:
def __init__(self, model_name="Qwen/Qwen3-235B-A22B"):
self.tokenizer = AutoTokenizer.from_pretrained(model_name)
self.model = AutoModelForCausalLM.from_pretrained(model_name)
self.history = []
def generate_response(self, user_input):
messages = self.history + [{"role": "user", "content": user_input}]
text = self.tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
inputs = self.tokenizer(text, return_tensors="pt")
response_ids = self.model.generate(**inputs, max_new_tokens=32768)[0][len(inputs.input_ids[0]):].tolist()
response = self.tokenizer.decode(response_ids, skip_special_tokens=True)
# Update history
self.history.append({"role": "user", "content": user_input})
self.history.append({"role": "assistant", "content": response})
return response
# Example Usage
if __name__ == "__main__":
chatbot = QwenChatbot()
# First input (without /think or /no_think tags, thinking mode is enabled by default)
user_input_1 = "How many r's in strawberries?"
print(f"User: {user_input_1}")
response_1 = chatbot.generate_response(user_input_1)
print(f"Bot: {response_1}")
print("----------------------")
# Second input with /no_think
user_input_2 = "Then, how many r's in blueberries? /no_think"
print(f"User: {user_input_2}")
response_2 = chatbot.generate_response(user_input_2)
print(f"Bot: {response_2}")
print("----------------------")
# Third input with /think
user_input_3 = "Really? /think"
print(f"User: {user_input_3}")
response_3 = chatbot.generate_response(user_input_3)
print(f"Bot: {response_3}")
```
> [!NOTE]
> For API compatibility, when `enable_thinking=True`, regardless of whether the user uses `/think` or `/no_think`, the model will always output a block wrapped in `<think>...</think>`. However, the content inside this block may be empty if thinking is disabled.
> When `enable_thinking=False`, the soft switches are not valid. Regardless of any `/think` or `/no_think` tags input by the user, the model will not generate think content and will not include a `<think>...</think>` block.
## Agentic Use
Qwen3 excels in tool calling capabilities. We recommend using [Qwen-Agent](https://github.com/QwenLM/Qwen-Agent) to make the best use of agentic ability of Qwen3. Qwen-Agent encapsulates tool-calling templates and tool-calling parsers internally, greatly reducing coding complexity.
To define the available tools, you can use the MCP configuration file, use the integrated tool of Qwen-Agent, or integrate other tools by yourself.
```python
from qwen_agent.agents import Assistant
# Define LLM
llm_cfg = {
'model': 'Qwen3-235B-A22B',
# Use the endpoint provided by Alibaba Model Studio:
# 'model_type': 'qwen_dashscope',
# 'api_key': os.getenv('DASHSCOPE_API_KEY'),
# Use a custom endpoint compatible with OpenAI API:
'model_server': 'http://localhost:8000/v1', # api_base
'api_key': 'EMPTY',
# Other parameters:
# 'generate_cfg': {
# # Add: When the response content is `<think>this is the thought</think>this is the answer;
# # Do not add: When the response has been separated by reasoning_content and content.
# 'thought_in_content': True,
# },
}
# Define Tools
tools = [
{'mcpServers': { # You can specify the MCP configuration file
'time': {
'command': 'uvx',
'args': ['mcp-server-time', '--local-timezone=Asia/Shanghai']
},
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
},
'code_interpreter', # Built-in tools
]
# Define Agent
bot = Assistant(llm=llm_cfg, function_list=tools)
# Streaming generation
messages = [{'role': 'user', 'content': 'https://qwenlm.github.io/blog/ Introduce the latest developments of Qwen'}]
for responses in bot.run(messages=messages):
pass
print(responses)
```
## Processing Long Texts
Qwen3 natively supports context lengths of up to 32,768 tokens. For conversations where the total length (including both input and output) significantly exceeds this limit, we recommend using RoPE scaling techniques to handle long texts effectively. We have validated the model's performance on context lengths of up to 131,072 tokens using the [YaRN](https://arxiv.org/abs/2309.00071) method.
YaRN is currently supported by several inference frameworks, e.g., `transformers` and `llama.cpp` for local use, `vllm` and `sglang` for deployment. In general, there are two approaches to enabling YaRN for supported frameworks:
- Modifying the model files:
In the `config.json` file, add the `rope_scaling` fields:
```json
{
...,
"rope_scaling": {
"type": "yarn",
"factor": 4.0,
"original_max_position_embeddings": 32768
}
}
```
For `llama.cpp`, you need to regenerate the GGUF file after the modification.
- Passing command line arguments:
For `vllm`, you can use
```shell
vllm serve ... --rope-scaling '{"type":"yarn","factor":4.0,"original_max_position_embeddings":32768}' --max-model-len 131072
```
For `sglang`, you can use
```shell
python -m sglang.launch_server ... --json-model-override-args '{"rope_scaling":{"type":"yarn","factor":4.0,"original_max_position_embeddings":32768}}'
```
For `llama-server` from `llama.cpp`, you can use
```shell
llama-server ... --rope-scaling yarn --rope-scale 4 --yarn-orig-ctx 32768
```
> [!IMPORTANT]
> If you encounter the following warning
> ```
> Unrecognized keys in `rope_scaling` for 'rope_type'='yarn': {'original_max_position_embeddings'}
> ```
> please upgrade `transformers>=4.51.0`.
> [!NOTE]
> All the notable open-source frameworks implement static YaRN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts.**
> We advise adding the `rope_scaling` configuration only when processing long contexts is required.
> It is also recommended to modify the `factor` as needed. For example, if the typical context length for your application is 65,536 tokens, it would be better to set `factor` as 2.0.
> [!NOTE]
> The default `max_position_embeddings` in `config.json` is set to 40,960. This allocation includes reserving 32,768 tokens for outputs and 8,192 tokens for typical prompts, which is sufficient for most scenarios involving short text processing. If the average context length does not exceed 32,768 tokens, we do not recommend enabling YaRN in this scenario, as it may potentially degrade model performance.
> [!TIP]
> The endpoint provided by Alibaba Model Studio supports dynamic YaRN by default and no extra configuration is needed.
## Best Practices
To achieve optimal performance, we recommend the following settings:
1. **Sampling Parameters**:
- For thinking mode (`enable_thinking=True`), use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0`. **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions.
- For non-thinking mode (`enable_thinking=False`), we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`.
- For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
2. **Adequate Output Length**: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 38,912 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
3. **Standardize Output Format**: We recommend using prompts to standardize model outputs when benchmarking.
- **Math Problems**: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
- **Multiple-Choice Questions**: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the `answer` field with only the choice letter, e.g., `"answer": "C"`."
4. **No Thinking Content in History**: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
### Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen3,
title = {Qwen3},
url = {https://qwenlm.github.io/blog/qwen3/},
author = {Qwen Team},
month = {April},
year = {2025}
}
```
|
pedalnomica/Qwen3-30B-A3B
|
pedalnomica
| 2025-04-28T21:43:44Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen3_moe",
"text-generation",
"conversational",
"arxiv:2309.00071",
"base_model:Qwen/Qwen3-30B-A3B-Base",
"base_model:finetune:Qwen/Qwen3-30B-A3B-Base",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T21:43:22Z |
---
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-30B-A3B/blob/main/LICENSE
pipeline_tag: text-generation
base_model:
- Qwen/Qwen3-30B-A3B-Base
---
# Qwen3-30B-A3B
<a href="https://chat.qwen.ai/" target="_blank" style="margin: 2px;">
<img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
</a>
## Qwen3 Highlights
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
- **Uniquely support of seamless switching between thinking mode** (for complex logical reasoning, math, and coding) and **non-thinking mode** (for efficient, general-purpose dialogue) **within single model**, ensuring optimal performance across various scenarios.
- **Significantly enhancement in its reasoning capabilities**, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
- **Superior human preference alignment**, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
- **Expertise in agent capabilities**, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
- **Support of 100+ languages and dialects** with strong capabilities for **multilingual instruction following** and **translation**.
## Model Overview
**Qwen3-30B-A3B** has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Number of Parameters: 30.5B in total and 3.3B activated
- Number of Paramaters (Non-Embedding): 29.9B
- Number of Layers: 48
- Number of Attention Heads (GQA): 32 for Q and 4 for KV
- Number of Experts: 128
- Number of Activated Experts: 8
- Context Length: 32,768 natively and [131,072 tokens with YaRN](#processing-long-texts).
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [blog](https://qwenlm.github.io/blog/qwen3/), [GitHub](https://github.com/QwenLM/Qwen3), and [Documentation](https://qwen.readthedocs.io/en/latest/).
## Quickstart
The code of Qwen3-MoE has been in the latest Hugging Face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.51.0`, you will encounter the following error:
```
KeyError: 'qwen3_moe'
```
The following contains a code snippet illustrating how to use the model generate content based on given inputs.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen3-30B-A3B"
# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
# conduct text completion
generated_ids = model.generate(
**model_inputs,
max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
# parsing thinking content
try:
# rindex finding 151668 (</think>)
index = len(output_ids) - output_ids[::-1].index(151668)
except ValueError:
index = 0
thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
print("thinking content:", thinking_content)
print("content:", content)
```
For deployment, you can use `sglang>=0.4.6.post1` or `vllm>=0.8.4` or to create an OpenAI-compatible API endpoint:
- SGLang:
```shell
python -m sglang.launch_server --model-path Qwen/Qwen3-30B-A3B --reasoning-parser qwen3
```
- vLLM:
```shell
vllm serve Qwen/Qwen3-30B-A3B --enable-reasoning --reasoning-parser deepseek_r1
```
For local use, applications such as llama.cpp, Ollama, LMStudio, and MLX-LM have also supported Qwen3.
## Switching Between Thinking and Non-Thinking Mode
> [!TIP]
> The `enable_thinking` switch is also available in APIs created by SGLang and vLLM.
> Please refer to our documentation for [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) and [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) users.
### `enable_thinking=True`
By default, Qwen3 has thinking capabilities enabled, similar to QwQ-32B. This means the model will use its reasoning abilities to enhance the quality of generated responses. For example, when explicitly setting `enable_thinking=True` or leaving it as the default value in `tokenizer.apply_chat_template`, the model will engage its thinking mode.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # True is the default value for enable_thinking
)
```
In this mode, the model will generate think content wrapped in a `<think>...</think>` block, followed by the final response.
> [!NOTE]
> For thinking mode, use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0` (the default setting in `generation_config.json`). **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### `enable_thinking=False`
We provide a hard switch to strictly disable the model's thinking behavior, aligning its functionality with the previous Qwen2.5-Instruct models. This mode is particularly useful in scenarios where disabling thinking is essential for enhancing efficiency.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=False # Setting enable_thinking=False disables thinking mode
)
```
In this mode, the model will not generate any think content and will not include a `<think>...</think>` block.
> [!NOTE]
> For non-thinking mode, we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### Advanced Usage: Switching Between Thinking and Non-Thinking Modes via User Input
We provide a soft switch mechanism that allows users to dynamically control the model's behavior when `enable_thinking=True`. Specifically, you can add `/think` and `/no_think` to user prompts or system messages to switch the model's thinking mode from turn to turn. The model will follow the most recent instruction in multi-turn conversations.
Here is an example of a multi-turn conversation:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
class QwenChatbot:
def __init__(self, model_name="Qwen/Qwen3-30B-A3B"):
self.tokenizer = AutoTokenizer.from_pretrained(model_name)
self.model = AutoModelForCausalLM.from_pretrained(model_name)
self.history = []
def generate_response(self, user_input):
messages = self.history + [{"role": "user", "content": user_input}]
text = self.tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
inputs = self.tokenizer(text, return_tensors="pt")
response_ids = self.model.generate(**inputs, max_new_tokens=32768)[0][len(inputs.input_ids[0]):].tolist()
response = self.tokenizer.decode(response_ids, skip_special_tokens=True)
# Update history
self.history.append({"role": "user", "content": user_input})
self.history.append({"role": "assistant", "content": response})
return response
# Example Usage
if __name__ == "__main__":
chatbot = QwenChatbot()
# First input (without /think or /no_think tags, thinking mode is enabled by default)
user_input_1 = "How many r's in strawberries?"
print(f"User: {user_input_1}")
response_1 = chatbot.generate_response(user_input_1)
print(f"Bot: {response_1}")
print("----------------------")
# Second input with /no_think
user_input_2 = "Then, how many r's in blueberries? /no_think"
print(f"User: {user_input_2}")
response_2 = chatbot.generate_response(user_input_2)
print(f"Bot: {response_2}")
print("----------------------")
# Third input with /think
user_input_3 = "Really? /think"
print(f"User: {user_input_3}")
response_3 = chatbot.generate_response(user_input_3)
print(f"Bot: {response_3}")
```
> [!NOTE]
> For API compatibility, when `enable_thinking=True`, regardless of whether the user uses `/think` or `/no_think`, the model will always output a block wrapped in `<think>...</think>`. However, the content inside this block may be empty if thinking is disabled.
> When `enable_thinking=False`, the soft switches are not valid. Regardless of any `/think` or `/no_think` tags input by the user, the model will not generate think content and will not include a `<think>...</think>` block.
## Agentic Use
Qwen3 excels in tool calling capabilities. We recommend using [Qwen-Agent](https://github.com/QwenLM/Qwen-Agent) to make the best use of agentic ability of Qwen3. Qwen-Agent encapsulates tool-calling templates and tool-calling parsers internally, greatly reducing coding complexity.
To define the available tools, you can use the MCP configuration file, use the integrated tool of Qwen-Agent, or integrate other tools by yourself.
```python
from qwen_agent.agents import Assistant
# Define LLM
llm_cfg = {
'model': 'Qwen3-30B-A3B',
# Use the endpoint provided by Alibaba Model Studio:
# 'model_type': 'qwen_dashscope',
# 'api_key': os.getenv('DASHSCOPE_API_KEY'),
# Use a custom endpoint compatible with OpenAI API:
'model_server': 'http://localhost:8000/v1', # api_base
'api_key': 'EMPTY',
# Other parameters:
# 'generate_cfg': {
# # Add: When the response content is `<think>this is the thought</think>this is the answer;
# # Do not add: When the response has been separated by reasoning_content and content.
# 'thought_in_content': True,
# },
}
# Define Tools
tools = [
{'mcpServers': { # You can specify the MCP configuration file
'time': {
'command': 'uvx',
'args': ['mcp-server-time', '--local-timezone=Asia/Shanghai']
},
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
},
'code_interpreter', # Built-in tools
]
# Define Agent
bot = Assistant(llm=llm_cfg, function_list=tools)
# Streaming generation
messages = [{'role': 'user', 'content': 'https://qwenlm.github.io/blog/ Introduce the latest developments of Qwen'}]
for responses in bot.run(messages=messages):
pass
print(responses)
```
## Processing Long Texts
Qwen3 natively supports context lengths of up to 32,768 tokens. For conversations where the total length (including both input and output) significantly exceeds this limit, we recommend using RoPE scaling techniques to handle long texts effectively. We have validated the model's performance on context lengths of up to 131,072 tokens using the [YaRN](https://arxiv.org/abs/2309.00071) method.
YaRN is currently supported by several inference frameworks, e.g., `transformers` and `llama.cpp` for local use, `vllm` and `sglang` for deployment. In general, there are two approaches to enabling YaRN for supported frameworks:
- Modifying the model files:
In the `config.json` file, add the `rope_scaling` fields:
```json
{
...,
"rope_scaling": {
"type": "yarn",
"factor": 4.0,
"original_max_position_embeddings": 32768
}
}
```
For `llama.cpp`, you need to regenerate the GGUF file after the modification.
- Passing command line arguments:
For `vllm`, you can use
```shell
vllm serve ... --rope-scaling '{"type":"yarn","factor":4.0,"original_max_position_embeddings":32768}' --max-model-len 131072
```
For `sglang`, you can use
```shell
python -m sglang.launch_server ... --json-model-override-args '{"rope_scaling":{"type":"yarn","factor":4.0,"original_max_position_embeddings":32768}}'
```
For `llama-server` from `llama.cpp`, you can use
```shell
llama-server ... --rope-scaling yarn --rope-scale 4 --yarn-orig-ctx 32768
```
> [!IMPORTANT]
> If you encounter the following warning
> ```
> Unrecognized keys in `rope_scaling` for 'rope_type'='yarn': {'original_max_position_embeddings'}
> ```
> please upgrade `transformers>=4.51.0`.
> [!NOTE]
> All the notable open-source frameworks implement static YaRN, which means the scaling factor remains constant regardless of input length, **potentially impacting performance on shorter texts.**
> We advise adding the `rope_scaling` configuration only when processing long contexts is required.
> It is also recommended to modify the `factor` as needed. For example, if the typical context length for your application is 65,536 tokens, it would be better to set `factor` as 2.0.
> [!NOTE]
> The default `max_position_embeddings` in `config.json` is set to 40,960. This allocation includes reserving 32,768 tokens for outputs and 8,192 tokens for typical prompts, which is sufficient for most scenarios involving short text processing. If the average context length does not exceed 32,768 tokens, we do not recommend enabling YaRN in this scenario, as it may potentially degrade model performance.
> [!TIP]
> The endpoint provided by Alibaba Model Studio supports dynamic YaRN by default and no extra configuration is needed.
## Best Practices
To achieve optimal performance, we recommend the following settings:
1. **Sampling Parameters**:
- For thinking mode (`enable_thinking=True`), use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0`. **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions.
- For non-thinking mode (`enable_thinking=False`), we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`.
- For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
2. **Adequate Output Length**: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 38,912 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
3. **Standardize Output Format**: We recommend using prompts to standardize model outputs when benchmarking.
- **Math Problems**: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
- **Multiple-Choice Questions**: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the `answer` field with only the choice letter, e.g., `"answer": "C"`."
4. **No Thinking Content in History**: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
### Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen3,
title = {Qwen3},
url = {https://qwenlm.github.io/blog/qwen3/},
author = {Qwen Team},
month = {April},
year = {2025}
}
```
|
kokovova/6aea4db7-12d6-4b5f-a345-b9577ffa5807
|
kokovova
| 2025-04-28T21:42:59Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"gemma2",
"axolotl",
"generated_from_trainer",
"base_model:princeton-nlp/gemma-2-9b-it-SimPO",
"base_model:adapter:princeton-nlp/gemma-2-9b-it-SimPO",
"license:mit",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T21:34:57Z |
---
library_name: peft
license: mit
base_model: princeton-nlp/gemma-2-9b-it-SimPO
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 6aea4db7-12d6-4b5f-a345-b9577ffa5807
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
adapter: lora
base_model: princeton-nlp/gemma-2-9b-it-SimPO
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 25962db5e0acc41e_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/25962db5e0acc41e_train_data.json
type:
field_instruction: topic
field_output: argument
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: kokovova/6aea4db7-12d6-4b5f-a345-b9577ffa5807
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/25962db5e0acc41e_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 8765f52f-03cf-464d-82e4-3ffbc452aff3
wandb_project: s56-4
wandb_run: your_name
wandb_runid: 8765f52f-03cf-464d-82e4-3ffbc452aff3
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 6aea4db7-12d6-4b5f-a345-b9577ffa5807
This model is a fine-tuned version of [princeton-nlp/gemma-2-9b-it-SimPO](https://huggingface.co/princeton-nlp/gemma-2-9b-it-SimPO) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4266
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 2.531 | 0.0571 | 200 | 2.4266 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
infogep/54ae5303-ca7f-4e4a-bd0a-c76cc5ea29e6
|
infogep
| 2025-04-28T21:42:27Z | 0 | 0 |
peft
|
[
"peft",
"safetensors",
"gemma2",
"axolotl",
"generated_from_trainer",
"base_model:princeton-nlp/gemma-2-9b-it-SimPO",
"base_model:adapter:princeton-nlp/gemma-2-9b-it-SimPO",
"license:mit",
"4-bit",
"bitsandbytes",
"region:us"
] | null | 2025-04-28T21:31:22Z |
---
library_name: peft
license: mit
base_model: princeton-nlp/gemma-2-9b-it-SimPO
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 54ae5303-ca7f-4e4a-bd0a-c76cc5ea29e6
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
<details><summary>See axolotl config</summary>
axolotl version: `0.4.1`
```yaml
absolute_data_files: false
adapter: lora
base_model: princeton-nlp/gemma-2-9b-it-SimPO
bf16: true
chat_template: llama3
dataset_prepared_path: /workspace/axolotl
datasets:
- data_files:
- 25962db5e0acc41e_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/25962db5e0acc41e_train_data.json
type:
field_instruction: topic
field_output: argument
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 1
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_clipping: 0.5
group_by_length: false
hub_model_id: infogep/54ae5303-ca7f-4e4a-bd0a-c76cc5ea29e6
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 5.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 8
mixed_precision: bf16
mlflow_experiment_name: /tmp/25962db5e0acc41e_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 1
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 8765f52f-03cf-464d-82e4-3ffbc452aff3
wandb_project: s56-30
wandb_run: your_name
wandb_runid: 8765f52f-03cf-464d-82e4-3ffbc452aff3
warmup_steps: 5
weight_decay: 0.01
xformers_attention: true
```
</details><br>
# 54ae5303-ca7f-4e4a-bd0a-c76cc5ea29e6
This model is a fine-tuned version of [princeton-nlp/gemma-2-9b-it-SimPO](https://huggingface.co/princeton-nlp/gemma-2-9b-it-SimPO) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 2.4270
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 5
- training_steps: 200
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 2.5264 | 0.0571 | 200 | 2.4270 |
### Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1
|
pedalnomica/Qwen3-0.6B
|
pedalnomica
| 2025-04-28T21:41:24Z | 0 | 0 |
transformers
|
[
"transformers",
"safetensors",
"qwen3",
"text-generation",
"conversational",
"base_model:Qwen/Qwen3-0.6B-Base",
"base_model:finetune:Qwen/Qwen3-0.6B-Base",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] |
text-generation
| 2025-04-28T21:41:23Z |
---
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-0.6B/blob/main/LICENSE
pipeline_tag: text-generation
base_model:
- Qwen/Qwen3-0.6B-Base
---
# Qwen3-0.6B
<a href="https://chat.qwen.ai/" target="_blank" style="margin: 2px;">
<img alt="Chat" src="https://img.shields.io/badge/%F0%9F%92%9C%EF%B8%8F%20Qwen%20Chat%20-536af5" style="display: inline-block; vertical-align: middle;"/>
</a>
## Qwen3 Highlights
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
- **Uniquely support of seamless switching between thinking mode** (for complex logical reasoning, math, and coding) and **non-thinking mode** (for efficient, general-purpose dialogue) **within single model**, ensuring optimal performance across various scenarios.
- **Significantly enhancement in its reasoning capabilities**, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
- **Superior human preference alignment**, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
- **Expertise in agent capabilities**, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
- **Support of 100+ languages and dialects** with strong capabilities for **multilingual instruction following** and **translation**.
## Model Overview
**Qwen3-0.6B** has the following features:
- Type: Causal Language Models
- Training Stage: Pretraining & Post-training
- Number of Parameters: 0.6B
- Number of Paramaters (Non-Embedding): 0.44B
- Number of Layers: 28
- Number of Attention Heads (GQA): 16 for Q and 8 for KV
- Context Length: 32,768
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our [blog](https://qwenlm.github.io/blog/qwen3/), [GitHub](https://github.com/QwenLM/Qwen3), and [Documentation](https://qwen.readthedocs.io/en/latest/).
> [!TIP]
> If you encounter significant endless repetitions, please refer to the [Best Practices](#best-practices) section for optimal sampling parameters, and set the ``presence_penalty`` to 1.5.
## Quickstart
The code of Qwen3 has been in the latest Hugging Face `transformers` and we advise you to use the latest version of `transformers`.
With `transformers<4.51.0`, you will encounter the following error:
```
KeyError: 'qwen3'
```
The following contains a code snippet illustrating how to use the model generate content based on given inputs.
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "Qwen/Qwen3-0.6B"
# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"
)
# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # Switches between thinking and non-thinking modes. Default is True.
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
# conduct text completion
generated_ids = model.generate(
**model_inputs,
max_new_tokens=32768
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()
# parsing thinking content
try:
# rindex finding 151668 (</think>)
index = len(output_ids) - output_ids[::-1].index(151668)
except ValueError:
index = 0
thinking_content = tokenizer.decode(output_ids[:index], skip_special_tokens=True).strip("\n")
content = tokenizer.decode(output_ids[index:], skip_special_tokens=True).strip("\n")
print("thinking content:", thinking_content)
print("content:", content)
```
For deployment, you can use `sglang>=0.4.6.post1` or `vllm>=0.8.4` or to create an OpenAI-compatible API endpoint:
- SGLang:
```shell
python -m sglang.launch_server --model-path Qwen/Qwen3-0.6B --reasoning-parser qwen3
```
- vLLM:
```shell
vllm serve Qwen/Qwen3-0.6B --enable-reasoning --reasoning-parser deepseek_r1
```
For local use, applications such as llama.cpp, Ollama, LMStudio, and MLX-LM have also supported Qwen3.
## Switching Between Thinking and Non-Thinking Mode
> [!TIP]
> The `enable_thinking` switch is also available in APIs created by SGLang and vLLM.
> Please refer to our documentation for [SGLang](https://qwen.readthedocs.io/en/latest/deployment/sglang.html#thinking-non-thinking-modes) and [vLLM](https://qwen.readthedocs.io/en/latest/deployment/vllm.html#thinking-non-thinking-modes) users.
### `enable_thinking=True`
By default, Qwen3 has thinking capabilities enabled, similar to QwQ-32B. This means the model will use its reasoning abilities to enhance the quality of generated responses. For example, when explicitly setting `enable_thinking=True` or leaving it as the default value in `tokenizer.apply_chat_template`, the model will engage its thinking mode.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=True # True is the default value for enable_thinking
)
```
In this mode, the model will generate think content wrapped in a `<think>...</think>` block, followed by the final response.
> [!NOTE]
> For thinking mode, use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0` (the default setting in `generation_config.json`). **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### `enable_thinking=False`
We provide a hard switch to strictly disable the model's thinking behavior, aligning its functionality with the previous Qwen2.5-Instruct models. This mode is particularly useful in scenarios where disabling thinking is essential for enhancing efficiency.
```python
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=False # Setting enable_thinking=False disables thinking mode
)
```
In this mode, the model will not generate any think content and will not include a `<think>...</think>` block.
> [!NOTE]
> For non-thinking mode, we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`. For more detailed guidance, please refer to the [Best Practices](#best-practices) section.
### Advanced Usage: Switching Between Thinking and Non-Thinking Modes via User Input
We provide a soft switch mechanism that allows users to dynamically control the model's behavior when `enable_thinking=True`. Specifically, you can add `/think` and `/no_think` to user prompts or system messages to switch the model's thinking mode from turn to turn. The model will follow the most recent instruction in multi-turn conversations.
Here is an example of a multi-turn conversation:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
class QwenChatbot:
def __init__(self, model_name="Qwen/Qwen3-0.6B"):
self.tokenizer = AutoTokenizer.from_pretrained(model_name)
self.model = AutoModelForCausalLM.from_pretrained(model_name)
self.history = []
def generate_response(self, user_input):
messages = self.history + [{"role": "user", "content": user_input}]
text = self.tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
inputs = self.tokenizer(text, return_tensors="pt")
response_ids = self.model.generate(**inputs, max_new_tokens=32768)[0][len(inputs.input_ids[0]):].tolist()
response = self.tokenizer.decode(response_ids, skip_special_tokens=True)
# Update history
self.history.append({"role": "user", "content": user_input})
self.history.append({"role": "assistant", "content": response})
return response
# Example Usage
if __name__ == "__main__":
chatbot = QwenChatbot()
# First input (without /think or /no_think tags, thinking mode is enabled by default)
user_input_1 = "How many r's in strawberries?"
print(f"User: {user_input_1}")
response_1 = chatbot.generate_response(user_input_1)
print(f"Bot: {response_1}")
print("----------------------")
# Second input with /no_think
user_input_2 = "Then, how many r's in blueberries? /no_think"
print(f"User: {user_input_2}")
response_2 = chatbot.generate_response(user_input_2)
print(f"Bot: {response_2}")
print("----------------------")
# Third input with /think
user_input_3 = "Really? /think"
print(f"User: {user_input_3}")
response_3 = chatbot.generate_response(user_input_3)
print(f"Bot: {response_3}")
```
> [!NOTE]
> For API compatibility, when `enable_thinking=True`, regardless of whether the user uses `/think` or `/no_think`, the model will always output a block wrapped in `<think>...</think>`. However, the content inside this block may be empty if thinking is disabled.
> When `enable_thinking=False`, the soft switches are not valid. Regardless of any `/think` or `/no_think` tags input by the user, the model will not generate think content and will not include a `<think>...</think>` block.
## Agentic Use
Qwen3 excels in tool calling capabilities. We recommend using [Qwen-Agent](https://github.com/QwenLM/Qwen-Agent) to make the best use of agentic ability of Qwen3. Qwen-Agent encapsulates tool-calling templates and tool-calling parsers internally, greatly reducing coding complexity.
To define the available tools, you can use the MCP configuration file, use the integrated tool of Qwen-Agent, or integrate other tools by yourself.
```python
from qwen_agent.agents import Assistant
# Define LLM
llm_cfg = {
'model': 'Qwen3-0.6B',
# Use the endpoint provided by Alibaba Model Studio:
# 'model_type': 'qwen_dashscope',
# 'api_key': os.getenv('DASHSCOPE_API_KEY'),
# Use a custom endpoint compatible with OpenAI API:
'model_server': 'http://localhost:8000/v1', # api_base
'api_key': 'EMPTY',
# Other parameters:
# 'generate_cfg': {
# # Add: When the response content is `<think>this is the thought</think>this is the answer;
# # Do not add: When the response has been separated by reasoning_content and content.
# 'thought_in_content': True,
# },
}
# Define Tools
tools = [
{'mcpServers': { # You can specify the MCP configuration file
'time': {
'command': 'uvx',
'args': ['mcp-server-time', '--local-timezone=Asia/Shanghai']
},
"fetch": {
"command": "uvx",
"args": ["mcp-server-fetch"]
}
}
},
'code_interpreter', # Built-in tools
]
# Define Agent
bot = Assistant(llm=llm_cfg, function_list=tools)
# Streaming generation
messages = [{'role': 'user', 'content': 'https://qwenlm.github.io/blog/ Introduce the latest developments of Qwen'}]
for responses in bot.run(messages=messages):
pass
print(responses)
```
## Best Practices
To achieve optimal performance, we recommend the following settings:
1. **Sampling Parameters**:
- For thinking mode (`enable_thinking=True`), use `Temperature=0.6`, `TopP=0.95`, `TopK=20`, and `MinP=0`. **DO NOT use greedy decoding**, as it can lead to performance degradation and endless repetitions.
- For non-thinking mode (`enable_thinking=False`), we suggest using `Temperature=0.7`, `TopP=0.8`, `TopK=20`, and `MinP=0`.
- For supported frameworks, you can adjust the `presence_penalty` parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
2. **Adequate Output Length**: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 38,912 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
3. **Standardize Output Format**: We recommend using prompts to standardize model outputs when benchmarking.
- **Math Problems**: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
- **Multiple-Choice Questions**: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the `answer` field with only the choice letter, e.g., `"answer": "C"`."
4. **No Thinking Content in History**: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
### Citation
If you find our work helpful, feel free to give us a cite.
```
@misc{qwen3,
title = {Qwen3},
url = {https://qwenlm.github.io/blog/qwen3/},
author = {Qwen Team},
month = {April},
year = {2025}
}
```
|
NikolayKozloff/Qwen3-1.7B-Q8_0-GGUF
|
NikolayKozloff
| 2025-04-28T21:40:11Z | 0 | 1 |
transformers
|
[
"transformers",
"gguf",
"llama-cpp",
"gguf-my-repo",
"text-generation",
"base_model:Qwen/Qwen3-1.7B",
"base_model:quantized:Qwen/Qwen3-1.7B",
"license:apache-2.0",
"endpoints_compatible",
"region:us",
"conversational"
] |
text-generation
| 2025-04-28T21:39:59Z |
---
base_model: Qwen/Qwen3-1.7B
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-1.7B/blob/main/LICENSE
pipeline_tag: text-generation
tags:
- llama-cpp
- gguf-my-repo
---
# NikolayKozloff/Qwen3-1.7B-Q8_0-GGUF
This model was converted to GGUF format from [`Qwen/Qwen3-1.7B`](https://huggingface.co/Qwen/Qwen3-1.7B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
Refer to the [original model card](https://huggingface.co/Qwen/Qwen3-1.7B) for more details on the model.
## Use with llama.cpp
Install llama.cpp through brew (works on Mac and Linux)
```bash
brew install llama.cpp
```
Invoke the llama.cpp server or the CLI.
### CLI:
```bash
llama-cli --hf-repo NikolayKozloff/Qwen3-1.7B-Q8_0-GGUF --hf-file qwen3-1.7b-q8_0.gguf -p "The meaning to life and the universe is"
```
### Server:
```bash
llama-server --hf-repo NikolayKozloff/Qwen3-1.7B-Q8_0-GGUF --hf-file qwen3-1.7b-q8_0.gguf -c 2048
```
Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
Step 1: Clone llama.cpp from GitHub.
```
git clone https://github.com/ggerganov/llama.cpp
```
Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
```
cd llama.cpp && LLAMA_CURL=1 make
```
Step 3: Run inference through the main binary.
```
./llama-cli --hf-repo NikolayKozloff/Qwen3-1.7B-Q8_0-GGUF --hf-file qwen3-1.7b-q8_0.gguf -p "The meaning to life and the universe is"
```
or
```
./llama-server --hf-repo NikolayKozloff/Qwen3-1.7B-Q8_0-GGUF --hf-file qwen3-1.7b-q8_0.gguf -c 2048
```
|
mrafi/opt-1.3b
|
mrafi
| 2025-04-28T21:38:38Z | 0 | 0 |
transformers
|
[
"transformers",
"document-question-answering",
"en",
"dataset:microsoft/ms_marco",
"base_model:facebook/opt-1.3b",
"base_model:finetune:facebook/opt-1.3b",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] |
document-question-answering
| 2025-04-28T21:36:33Z |
---
license: apache-2.0
datasets:
- microsoft/ms_marco
language:
- en
base_model:
- facebook/opt-1.3b
pipeline_tag: document-question-answering
library_name: transformers
---
|
mlx-community/Qwen3-1.7B-8bit
|
mlx-community
| 2025-04-28T21:36:10Z | 0 | 0 |
mlx
|
[
"mlx",
"safetensors",
"qwen3",
"text-generation",
"conversational",
"base_model:Qwen/Qwen3-1.7B",
"base_model:quantized:Qwen/Qwen3-1.7B",
"license:apache-2.0",
"8-bit",
"region:us"
] |
text-generation
| 2025-04-28T21:34:08Z |
---
library_name: mlx
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3-1.7B/blob/main/LICENSE
pipeline_tag: text-generation
base_model: Qwen/Qwen3-1.7B
tags:
- mlx
---
# mlx-community/Qwen3-1.7B-8bit
This model [mlx-community/Qwen3-1.7B-8bit](https://huggingface.co/mlx-community/Qwen3-1.7B-8bit) was
converted to MLX format from [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B)
using mlx-lm version **0.24.0**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/Qwen3-1.7B-8bit")
prompt = "hello"
if tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)
```
|
TunisianCoder/zizou2547
|
TunisianCoder
| 2025-04-28T21:34:56Z | 0 | 0 | null |
[
"dataset:nvidia/OpenCodeReasoning",
"region:us"
] | null | 2025-04-28T21:34:37Z |
---
datasets:
- nvidia/OpenCodeReasoning
---
|
agoor97/onnx-models
|
agoor97
| 2025-04-28T21:34:24Z | 0 | 0 | null |
[
"onnx",
"model",
"quantization",
"RAG",
"chatbot",
"NLP",
"small-models",
"mobile-deployment",
"embedded-systems",
"en",
"license:apache-2.0",
"region:us"
] | null | 2025-04-22T01:27:22Z |
---
language: en
license: apache-2.0
tags:
- onnx
- model
- quantization
- RAG
- chatbot
- NLP
- small-models
- mobile-deployment
- embedded-systems
---
# π LLM to ONNX Converter
> Convert small language models to ONNX format with **guaranteed reliability** for RAG and chatbot applications on resource-constrained hardware.
## π Overview
This repository provides scripts to convert small language models to ONNX format and create INT8 quantized versions for efficient deployment on resource-constrained devices. Perfect for mobile applications, Unity game engines, and embedded systems.
## β
Tested Models
We've successfully tested the following models with example outputs:
| Model | Size | Quantized | Response Quality | Speed (sec) |
|-------|------|-----------|-----------------|-------------|
| Qwen-0.5B | 500M | β
| β Poor | 8.37 |
| Qwen-0.5B | 500M | β | β
Good | 15.69 |
| TinyLlama-1.1B | 1.1B | β
| β Poor | 10.15 |
| TinyLlama-1.1B | 1.1B | β | β
Good | 19.23 |
| Phi-1.5 | 1.3B | β | β
Good | 15.32 |
| Falcon-RW-1B | 1B | β | β
Good | 21.56 |
| GPT2-Medium | 355M | β
| β
Good | 6.27 |
| GPT2-Medium | 355M | β | β
Good | 12.77 |
| OPT-350M | 350M | β
| β
Good | 4.33 |
| OPT-350M | 350M | β | β
Good | 10.42 |
| Bloom-560M | 560M | β
| β Poor | 11.93 |
| Bloom-560M | 560M | β | β
Good | 34.38 |
## π Recommendations
Based on our testing:
1. **For best speed + quality:** OPT-350M (quantized) - fastest with good quality
2. **For best overall quality:** Phi-1.5 (non-quantized) - excellent responses
3. **For smallest size:** GPT2-Medium or OPT-350M (quantized) - small with good performance
## π© Key Findings
- Quantization provides ~2x speed improvement
- Smaller models (350-500M) quantize better than larger models (1B+)
- Some architectures (OPT, GPT2) handle quantization better than others
## π Repository Structure
```
onnx_models/
βββ bloom_onnx/
βββ bloom_onnx_quantized/
βββ falcon_onnx/
βββ gpt2_onnx/
βββ gpt2_onnx_quantized/
βββ opt_onnx/
βββ opt_onnx_quantized/
βββ phi_onnx/
βββ qwen_onnx/
βββ qwen_onnx_quantized/
βββ tinyllama_onnx/
βββ tinyllama_onnx_quantized/
```
## π Requirements
- Python 3.8+
- optimum
- onnxruntime
- transformers
- numpy
---------------
|
Subsets and Splits
Filtered Qwen2.5 Distill Models
Identifies specific configurations of models by filtering cards that contain 'distill', 'qwen2.5', '7b' while excluding certain base models and incorrect model ID patterns, uncovering unique model variants.
Filtered Model Cards Count
Finds the count of entries with specific card details that include 'distill', 'qwen2.5', '7b' but exclude certain base models, revealing valuable insights about the dataset's content distribution.
Filtered Distill Qwen 7B Models
Filters for specific card entries containing 'distill', 'qwen', and '7b', excluding certain strings and patterns, to identify relevant model configurations.
Filtered Qwen-7b Model Cards
The query performs a detailed filtering based on specific keywords and excludes certain entries, which could be useful for identifying a specific subset of cards but does not provide deeper insights or trends.
Filtered Qwen 7B Model Cards
The query filters for specific terms related to "distilled" or "distill", "qwen", and "7b" in the 'card' column but excludes certain base models, providing a limited set of entries for further inspection.
Qwen 7B Distilled Models
The query provides a basic filtering of records to find specific card names that include keywords related to distilled Qwen 7b models, excluding a particular base model, which gives limited insight but helps in focusing on relevant entries.
Qwen 7B Distilled Model Cards
The query filters data based on specific keywords in the modelId and card fields, providing limited insight primarily useful for locating specific entries rather than revealing broad patterns or trends.
Qwen 7B Distilled Models
Finds all entries containing the terms 'distilled', 'qwen', and '7b' in a case-insensitive manner, providing a filtered set of records but without deeper analysis.
Distilled Qwen 7B Models
The query filters for specific model IDs containing 'distilled', 'qwen', and '7b', providing a basic retrieval of relevant entries but without deeper analysis or insight.
Filtered Model Cards with Distill Qwen2.
Filters and retrieves records containing specific keywords in the card description while excluding certain phrases, providing a basic count of relevant entries.
Filtered Model Cards with Distill Qwen 7
The query filters specific variations of card descriptions containing 'distill', 'qwen', and '7b' while excluding a particular base model, providing limited but specific data retrieval.
Distill Qwen 7B Model Cards
The query filters and retrieves rows where the 'card' column contains specific keywords ('distill', 'qwen', and '7b'), providing a basic filter result that can help in identifying specific entries.