torch torchvision torchaudio matplotlib wurlitzer accelerate bitsandbytes https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu118torch1.12cxx11abiFALSE-cp310-cp310-linux_x86_64.whl optimum # auto-gptq gradio TTS transformers numpy openai-whisper geopy replicate langchain langchain-community text_generation python-dotenv pydantic-settings ollama plotly-express loguru