kanshan/minimind
1
0
Fork 0
mirror of https://github.com/jingyaogong/minimind.git synced 2026-01-13 19:57:20 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

[feat] add MNN support to README.

Browse Source
This commit is contained in:
yanxing 2025-11-10 21:56:27 +08:00
parent bf60bde8fb
commit 5959396096
2 changed files with 34 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
README.md
View File

@ -1799,6 +1799,23 @@ ollama run jingyaogong/minimind2 # 其他可选 minimind2-r1 / minimind2-small /
我是一个语言模型...
```
## <img src="https://github.com/alibaba/MNN/blob/master/doc/banner.png" height="28" style="vertical-align: middle;"/> [MNN](https://github.com/alibaba/MNN)
MNN是面向端侧的AI推理引擎，支持多种开源LLM模型推理，轻量化、高性能。
1. 模型转换
```
cd MNN/transformers/llm/export
# 导出4bit HQQ量化的MNN模型
python llmexport.py --path /path/to/MiniMind2/ --export mnn --hqq --dst_path MiniMind2-MNN
```
2. 在Mac或手机上测试
```
./llm_demo /path/to/MiniMind2-MNN/config.json prompt.txt
```
或者下载APP测试
> 以上三方框架的更多用法请参考对应官方文档😊
# 📌 Acknowledge

17
README_en.md
View File

@ -1728,6 +1728,23 @@ ollama run jingyaogong/minimind2 # Other options: minimind2-r1 / minimind2-small
I am a language model...
```
## <img src="https://github.com/alibaba/MNN/blob/master/doc/banner.png" height="28" style="vertical-align: middle;"/> [MNN](https://github.com/alibaba/MNN)
MNN is a lightweight, high-performance AI inference engine for on-device applications, supporting inference for various open-source LLM models.
1. **Model Conversion**
```
cd MNN/transformers/llm/export
# Export the 4-bit HQQ quantized MNN model
python llmexport.py --path /path/to/MiniMind2/ --export mnn --hqq --dst_path MiniMind2-MNN
```
2. **Test on a Mac or mobile phone**
```
./llm_demo /path/to/MiniMind2-MNN/config.json prompt.txt
```
Or download the app to test.
> For more usage of the above third-party frameworks, please refer to their official documentation 😊
# 📌 Acknowledge