---
title: DeepSeek-R1-0528-4bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/DeepSeek-R1-0528-4bit"
md_url: "https://www.modelscope.cn/models/mlx-community/DeepSeek-R1-0528-4bit.md"
repository: mlx-community/DeepSeek-R1-0528-4bit
last_updated: 2025-05-29
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - deepseek_v3
architectures:
  - DeepseekV3ForCausalLM
parameters: 104.9B
tensor_type:
  - BF16
  - U32
  - F32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
inference_backends:
  - "deploy_task text/emb"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 510
stars: 1
tags:
  - mlx
---

# DeepSeek-R1-0528-4bit

> DeepSeek-R1-0528-4bit - mlx-community 在 ModelScope 开源的模型。mlx-community/DeepSeek-R1-0528-4bit

mlx-community/DeepSeek-R1-0528-4bit 是 ModelScope 魔搭社区上的 104.9B 参数text-generation模型，可用 deploy_task text/emb、sglang 0.5.2、vllm 0.9.2 部署。

- **Repository**: mlx-community/DeepSeek-R1-0528-4bit
- **Tasks**: text-generation
- **Parameters**: 104.9B
- **Inference backends**: deploy_task text/emb, sglang 0.5.2, vllm 0.9.2
- **Tags**: mlx
- **Downloads**: 510
- **Stars**: 1
- **Last updated**: 2025-05-29

Source: https://www.modelscope.cn/models/mlx-community/DeepSeek-R1-0528-4bit

---

# mlx-community/DeepSeek-R1-0528-4bit

This model [mlx-community/DeepSeek-R1-0528-4bit](https://huggingface.co/mlx-community/DeepSeek-R1-0528-4bit/new/main) was
converted to MLX format from [deepseek-ai/DeepSeek-R1-0528](https://huggingface.co/deepseek-ai/DeepSeek-R1-0528)
using mlx-lm version **0.24.1**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/DeepSeek-R1-0528-4bit")

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)
```