mediapipe transformers>=4.46.3 diffusers==0.31.0 torchmetrics torchtyping tqdm einops==0.8.0 omegaconf==2.3.0 opencv-python av==13.1.0 gradio accelerate==1.1.1 clip @ https://github.com/openai/CLIP/archive/d50d76daa670286dd6cacf3bcd80b5e4823fc8e1.zip#sha256=b5842c25da441d6c581b53a5c60e0c2127ebafe0f746f8e15561a006c6c3be6a decord==0.6.0 diffusers==0.31.0 einops==0.8.0 gradio_client==1.4.3 imageio==2.36.0 imageio-ffmpeg==0.5.1 numpy==1.26.4 onnxruntime-gpu==1.20.1 open-clip-torch==2.29.0 opencv-contrib-python==4.10.0.84 Pillow>=10.4.0 scikit-image==0.24.0 scikit-learn==1.5.2 scipy==1.14.1 torchdiffeq==0.2.5 torchsde==0.2.6 mlflow==2.18.0 controlnet-aux==0.0.9 ffmpeg-python soundfile mediapipe decord IPython scenedetect moviepy==1.0.3 huggingface_hub==0.26.2 torchao