Instructions to use a-hamdi/NGILlama3-merged with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use a-hamdi/NGILlama3-merged with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="a-hamdi/NGILlama3-merged")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("a-hamdi/NGILlama3-merged")
model = AutoModelForCausalLM.from_pretrained("a-hamdi/NGILlama3-merged", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use a-hamdi/NGILlama3-merged with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "a-hamdi/NGILlama3-merged"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "a-hamdi/NGILlama3-merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/a-hamdi/NGILlama3-merged

SGLang

How to use a-hamdi/NGILlama3-merged with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "a-hamdi/NGILlama3-merged" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "a-hamdi/NGILlama3-merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "a-hamdi/NGILlama3-merged" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "a-hamdi/NGILlama3-merged",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use a-hamdi/NGILlama3-merged with Docker Model Runner:
```
docker model run hf.co/a-hamdi/NGILlama3-merged
```

NGI_LLama3_8b

Model Description

Model Description: unsloth/llama-3-8b-bnb-4bit

Model Card Title

NGI_LLama3_8b

Model Type

text-generation

Model Details

Dataset: Custom News dataset

Training Data Source: [Please provide the source]

Training Data Size: 840,000

Number of Training Steps: 210,000

Model Architecture: LLaMA

Technical Specifications

Framework: PyTorch

Hyperparameters:

Learning Rate: 5e-5
Batch Size: 4
Optimizer: adamw_8bit

Evaluation Metrics:

Metric Name: [Accuracy]
Score: [0.9981]
Metric Name: [F1 score]
Score: [0.9980]
Metric Name: [Precision]
Score: [0.9963]
Metric Name: [Recall]
Score: [0.9996]

Hardware Used:

Number of GPUs/CPUs: 1 GPU
Type of GPUs/CPUs: [NVIDIA RTX A4000 ]

Training Duration: [24 days]

Usage

Intended Use Cases: [Please provide the use cases]

Example Input: [Please provide an example input]

Example Output: [Please provide an example output]

Licensing and Access

License Type: [apache-2.0]

Model Access: [Public/Private]

Additional Information

Contributors: [Please provide the contributors]

Citations and References: [Please provide any citations or references]

Contact Information: [Please provide contact information]

Repository URL: [Please provide the repository URL, if any]

Additional Comments or Notes: [Please provide any additional comments or notes]

Downloads last month: 11

Safetensors

Model size

8B params

Tensor type

BF16

a-hamdi
/

NGILlama3-merged

NGI_LLama3_8b

Model Description

Model Card Title

Tags

Model Type

Model Details

Technical Specifications

Usage

Licensing and Access

Additional Information