LLM跑在边缘芯片上

1. 模型工具，转换，加载等等
   1.  Distributed Llama   https://github.com/b4rtaz/distributed-llama?tab=readme-ov-file
2. 语音模型
   1. https://github.com/k2-fsa/sherpa-onnx
   2. 小智生态 https://github.com/xinnan-tech/xiaozhi-esp32-server