| # Core ML/AI packages (pinned for compatibility) | |
| torch==2.3.1 | |
| torchaudio==2.3.1 | |
| transformers==4.44.2 | |
| accelerate==0.33.0 | |
| # Audio processing (required by MiniCPM-o 2.6) | |
| librosa==0.10.1 | |
| soundfile==0.12.1 | |
| scipy==1.11.4 | |
| # TTS dependencies (required by MiniCPM-o 2.6) | |
| vector_quantize_pytorch==1.14.24 | |
| vocos==0.1.0 | |
| # Flash Attention (prebuilt wheel for torch 2.3.1 + Python 3.10) | |
| https://github.com/Dao-AILab/flash-attention/releases/download/v2.7.4.post1/flash_attn-2.7.4.post1+cu12torch2.3cxx11abiFALSE-cp310-cp310-linux_x86_64.whl | |
| # Gradio and Spaces | |
| gradio==5.35.0 | |
| spaces==0.37.1 | |
| # Video/Image processing | |
| opencv-python-headless==4.9.0.80 | |
| pillow==10.4.0 | |
| decord==0.6.0 | |
| moviepy==1.0.3 | |
| # Core Python packages | |
| numpy==1.24.4 | |
| ffmpeg-python==0.2.0 | |
| # Additional dependencies for video processing | |
| imageio==2.34.2 | |
| imageio-ffmpeg==0.5.1 |