---
title: llama3.1-8b-instruct-dpo-zh
canonical_url: "https://www.modelscope.cn/models/shareAI/llama3.1-8b-instruct-dpo-zh"
md_url: "https://www.modelscope.cn/models/shareAI/llama3.1-8b-instruct-dpo-zh.md"
repository: shareAI/llama3.1-8b-instruct-dpo-zh
chinese_name: "llama3.1-8b instruct 中文DPO对话模型"
last_updated: 2024-07-25
license: "Apache License 2.0"
pipeline_tag: chatbot
tasks:
  - chatbot
model_type:
  - llama
architectures:
  - LlamaForCausalLM
parameters: 8.0B
tensor_type:
  - F16
library_name:
  - safetensors
  - pytorch
frameworks:
  - Pytorch
language:
  - cn
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 3714
stars: 15
tags:
  - RL-tuned
---

# llama3.1-8b-instruct-dpo-zh

> llama3.1-8b-instruct-dpo-zh - shareAI 在 ModelScope 开源的模型。llama3.1-8b-instruct 中文DPO版 模型介绍 Github：llama3中文仓库 像原版instruct一样，喜欢用有趣中文和表情符号回答问题。

shareAI/llama3.1-8b-instruct-dpo-zh 是 ModelScope 魔搭社区上的 8.0B 参数chatbot模型，采用 Apache License 2.0 许可，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: shareAI/llama3.1-8b-instruct-dpo-zh
- **License**: Apache License 2.0
- **Tasks**: chatbot
- **Parameters**: 8.0B
- **Inference backends**: deploy_task text/emb, lmdeploy 0.9.1, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: RL-tuned
- **Downloads**: 3714
- **Stars**: 15
- **Last updated**: 2024-07-25

Source: https://www.modelscope.cn/models/shareAI/llama3.1-8b-instruct-dpo-zh

---

# llama3.1-8b-instruct 中文DPO版
### 模型介绍
Github：[llama3中文仓库](https://github.com/CrazyBoyM/llama3-Chinese-chat)  
像原版instruct一样，喜欢用有趣中文和表情符号回答问题。  


特点：偏好中文和emoji表情，且不损伤原instruct版模型能力。实测中文DPO版问答性能体验超过现在市面上任何llama3.1中文微调版 （微调会大面积破坏llama3.1原版能力，导致遗忘）  
![Alt text](https://modelscope.cn/api/v1/models/baicai003/Llama3-Chinese-instruct-DPO-beta0.5/repo?Revision=master&FilePath=image.png&View=true)

### 训练细节
DPO(beta 0.5) + lora rank128, alpha256 + 打开"lm_head", "input_layernorm", "post_attention_layernorm", "norm"层训练.  

### 模型部署
网页脚本文件：https://github.com/CrazyBoyM/llama3-Chinese-chat/blob/main/deploy/web_streamlit_for_instruct_v2.py  
已经在模型文件夹里内置了一份web.py，可以直接使用。
```
pip install streamlit
pip install transformers==4.40.1
streamlit run web.py ./llama3.1-8b-instruct-dpo-zh
```

### 模型下载
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('shareAI/llama3.1-8b-instruct-dpo-zh')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/shareAI/llama3.1-8b-instruct-dpo-zh.git
```