---
title: DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx
canonical_url: "https://www.modelscope.cn/models/mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx"
md_url: "https://www.modelscope.cn/models/mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx.md"
repository: mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx
last_updated: 2024-12-19
license: other
model_type:
  - deepseek_v2
architectures:
  - DeepseekV2ForCausalLM
parameters: 2.5B
tensor_type:
  - U32
  - F16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 1082
stars: 0
tags:
  - mlx
---

# DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx

> DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx - mlx-community 在 ModelScope 开源的模型。mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx

mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx 是 ModelScope 魔搭社区上的 2.5B 参数机器学习模型，采用 other 许可，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx
- **License**: other
- **Parameters**: 2.5B
- **Inference backends**: deploy_task text/emb, lmdeploy 0.9.1, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: mlx
- **Downloads**: 1082
- **Stars**: 0
- **Last updated**: 2024-12-19

Source: https://www.modelscope.cn/models/mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx

---

# mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx

The Model [mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx](https://huggingface.co/mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx) was converted to MLX format from [deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct](https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct) using mlx-lm version **0.16.0**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx")
response = generate(model, tokenizer, prompt="hello", verbose=True)
```