| # Core dependencies | |
| numpy | |
| einops | |
| librosa | |
| pydub | |
| torch | |
| torchaudio | |
| torchvision | |
| torchcodec | |
| torchdiffeq | |
| transformers>=4.54.0 | |
| # Audio / multimodal models | |
| audiobox_aesthetics | |
| # Git-based dependencies | |
| git+https://github.com/huggingface/datasets.git | |
| git+https://github.com/huggingface/peft.git | |
| git+https://github.com/facebookresearch/sam3.git | |
| git+https://github.com/huggingface/accelerate.git | |
| dacvae @ git+https://github.com/facebookresearch/dacvae.git | |
| imagebind @ git+https://github.com/facebookresearch/ImageBind.git | |
| laion-clap @ git+https://github.com/lematt1991/CLAP.git | |
| #perception-models @ git+https://github.com/facebookresearch/perception_models@unpin-deps | |
| perception-models@git+https://github.com/hx23840/perception_models@unpin-deps |