Spaces:
Running
Running
| # Core AI stack | |
| diffusers==0.34.0 | |
| transformers==4.53.1 | |
| tokenizers>=0.20.3 | |
| accelerate>=1.1.1 | |
| tqdm | |
| imageio | |
| imageio-ffmpeg | |
| einops | |
| sentencepiece | |
| open_clip_torch>=2.29.0 | |
| # Video & media | |
| moviepy==1.0.3 | |
| av | |
| ffmpeg-python | |
| pygame>=2.1.0 | |
| sounddevice>=0.4.0 | |
| soundfile | |
| mutagen | |
| pyloudnorm | |
| librosa==0.11.0 | |
| speechbrain==1.0.3 | |
| audio-separator==0.36.1 | |
| # UI & interaction | |
| gradio==5.29.0 | |
| dashscope | |
| loguru | |
| # Vision & segmentation | |
| opencv-python>=4.12.0.88 | |
| segment-anything | |
| rembg[gpu]==2.0.65 | |
| onnxruntime-gpu==1.22 | |
| decord | |
| timm | |
| insightface @ https://github.com/deepbeepmeep/insightface/raw/refs/heads/master/wheels/insightface-0.7.3-cp310-cp310-win_amd64.whl ; sys_platform == "win32" and python_version == "3.10" | |
| insightface==0.7.3 ; sys_platform == "linux" | |
| facexlib==0.3.0 | |
| # Config & orchestration | |
| omegaconf | |
| hydra-core | |
| easydict | |
| pydantic==2.10.6 | |
| # Math & modeling | |
| torchdiffeq>=0.2.5 | |
| tensordict>=0.6.1 | |
| mmgp==3.6.2 | |
| peft==0.15.0 | |
| matplotlib | |
| # Utilities | |
| ftfy | |
| piexif | |
| nvidia-ml-py | |
| misaki | |
| # Optional / commented out | |
| # transformers==4.46.3 # for llamallava pre-patch | |
| # rembg==2.0.65 # non-GPU fallback | |
| # huggingface_hub[hf_xet] # slows down everything | |
| # num2words | |
| # spacy | |
| torch | |
| torchvision | |
| torchaudio | |
| transformers | |
| diffusers | |
| accelerate | |
| gradio | |
| tqdm | |
| safetensors | |
| opencv-python | |
| numpy | |
| Pillow | |