# Open Interpreter with Llama 2 [GitHub - KillianLucas/open-interpreter: A natural language interface for computers](https://github.com/KillianLucas/open-interpreter/?tab=readme-ov-file) ```sh interpreter --local ``` ## Using LM Studio [LM Studio - Discover, download, and run local LLMs](https://lmstudio.ai/) ```sh sudo add-apt-repository universe sudo apt install libfuse2 wget https://releases.lmstudio.ai/linux/0.2.14/beta/LM_Studio-0.2.14-beta-1.AppImage chmod u+x LM_Studio-0.2.14-beta-1.AppImage ./LM_Studio-0.2.14-beta-1.AppImage #=> Missing X server or $DISPLAY ``` It needs GUI. ## Without LM Studio [Open-interpreterを、自前で立てたOpenAI互換サーバで使う。llama-cpp-pythonとFastChat。｜めぐチャンネル](https://note.com/ai_meg/n/nc0206327dd8f) ```sh CMAKE_ARGS="-DLLAMA_CUBLAS=on" FORCE_CMAKE=1 pip install 'llama-cpp-python[server]' ``` ```sh python -m llama_cpp.server --host 192.168.0.102 --model ../models/codellama-13b.Q5_K_M.gguf --n_gpu_layers 8000 ``` ```sh curl http://192:169.0.102:8000/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "gpt-3.5-turbo", "messages": [ { "role": "user", "content": "Hello." } ] }' ``` `config.yaml` ```yaml llm.api_key: dummy llm.api_base: http://192.168.0.102:8000/v1 ``` ```sh interpreter -cf config.yaml ```