README.md

---
language:
- multilingual
tags:
- code-generation
- transformers
- mlx
- mlx-my-repo
license: mit
base_model: Kwaipilot/KwaiCoder-DS-V2-Lite-Base
---

# vopash/KwaiCoder-DS-V2-Lite-Base-mlx-4Bit

The Model [vopash/KwaiCoder-DS-V2-Lite-Base-mlx-4Bit](https://huggingface.co/vopash/KwaiCoder-DS-V2-Lite-Base-mlx-4Bit) was converted to MLX format from [Kwaipilot/KwaiCoder-DS-V2-Lite-Base](https://huggingface.co/Kwaipilot/KwaiCoder-DS-V2-Lite-Base) using mlx-lm version **0.29.1**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("vopash/KwaiCoder-DS-V2-Lite-Base-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)
```