Spaces:
Running on Zero
Running on Zero
File size: 1,598 Bytes
aa72c2d d35d872 aa72c2d 971b83d aa72c2d d35d872 27ffbbc d35d872 aa72c2d | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 | gradio>=5.0,<6.0 spaces>=0.30.0 huggingface_hub>=0.27.0 psutil # used by app.py model status badge torch>=2.4.0 torchvision # HF Spaces ships torch==2.10.0; bare 'torchaudio' lets pip pick an older # wheel whose C++ ABI doesn't match → import crashes the whole audio code # path including ComfyUI-LTXVideo's audio VAE. Pin the matching version. torchaudio==2.10.0 numpy Pillow einops safetensors tqdm # ComfyUI core requirements (also pip-installed by setup.sh from comfyui/requirements.txt # locally, but Spaces won't run setup.sh — these are the deps ComfyUI itself needs) transformers>=4.50,<6 tokenizers sentencepiece av kornia spandrel torchsde scipy aiohttp pydantic pydantic-settings python-dotenv yarl PyOpenGL glfw SQLAlchemy alembic Mako networkx sympy mpmath blake3 comfyui-frontend-package>=1.42 comfyui-workflow-templates>=0.9 comfyui-embedded-docs # Custom-node deps that we discovered are required for the LTX 2.3 workflow # but aren't auto-installed by their respective nodes' requirements.txt. gguf # ComfyUI-GGUF (UnetLoaderGGUF) imageio_ffmpeg # ComfyUI-VideoHelperSuite (video write/read backend) opencv-python # ComfyUI_LayerStyle, multiple custom nodes matplotlib # comfyui_controlnet_aux dwpose / pose preprocessors scikit-image # comfyui_controlnet_aux dwpose hand keypoint detector onnxruntime # comfyui_controlnet_aux dwpose accelerator (CPU/CoreML on Mac, CUDA on Spaces) diffusers # ComfyUI-SeedVR2 (used during init even when the node isn't called) yt-dlp # ComfyUI-MediaMixer (init-time import) # Dev / test pytest>=8.0 pytest-asyncio>=0.23 ruff>=0.5 |