---
title: DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa
canonical_url: "https://www.modelscope.cn/models/YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa"
md_url: "https://www.modelscope.cn/models/YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa.md"
repository: YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa
last_updated: 2025-03-19
license: "Apache License 2.0"
pipeline_tag: text-generation
tasks:
  - text-generation
base_model:
  - deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
base_model_relation: adapter
parameters: 157.3M
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - Pytorch
downloads: 75
stars: 0
---

# DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa

> DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa - YanfeiSong 在 ModelScope 开源的模型。用于DeepSeek-R1-Distill-Qwen-32B美杜莎推测解码的辅助模型,拥有3个头,每个头一层.

YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa 是 ModelScope 魔搭社区上的 157.3M 参数text-generation模型，采用 Apache License 2.0 许可，基于 deepseek-ai/DeepSeek-R1-Distill-Qwen-32B 构建。

- **Repository**: YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa
- **License**: Apache License 2.0
- **Tasks**: text-generation
- **Parameters**: 157.3M
- **Base model**: deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
- **Downloads**: 75
- **Stars**: 0
- **Last updated**: 2025-03-19

Source: https://www.modelscope.cn/models/YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa

---

### 此为Medusa官方Setup在shareGPT上训练的Medusa头, 实际使用会加速50%~5%不等。
#### 您可以通过如下git clone命令，或者ModelScope SDK来下载模型

SDK下载
```bash
#安装ModelScope
pip install modelscope
```
```python
#SDK模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/YanfeiSong/DeepSeek-R1-Distill-Qwen-32B-3heads-Medusa.git
```

<p style="color: lightgrey;">如果您是本模型的贡献者，我们邀请您根据<a href="https://modelscope.cn/docs/ModelScope%E6%A8%A1%E5%9E%8B%E6%8E%A5%E5%85%A5%E6%B5%81%E7%A8%8B%E6%A6%82%E8%A7%88" style="color: lightgrey; text-decoration: underline;">模型贡献文档</a>，及时完善模型卡片内容。</p>