File size: 1,098 Bytes
fa801f2 f3eda0e fa801f2 f3eda0e fa801f2 f3eda0e fa801f2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 |
---
library_name: transformers
license: mit
datasets:
- ethicalabs/Kurtis-E1-SFT
language:
- en
base_model: ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct
pipeline_tag: text-generation
tags:
- mlx
---
# ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct-mlx-4Bit
The Model [ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct-mlx-4Bit](https://huggingface.co/ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct-mlx-4Bit) was converted to MLX format from [ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct](https://huggingface.co/ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct) using mlx-lm version **0.22.1**.
## Use with mlx
```bash
pip install mlx-lm
```
```python
from mlx_lm import load, generate
model, tokenizer = load("ethicalabs/Kurtis-E1.1-Qwen2.5-0.5B-Instruct-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)
```
|