character-studio / requirements.txt
mamungtai-sat's picture
Add Pose lock mode (ControlNet OpenPose) for SD1.5 (#13)
0e89ded
Raw
History Blame
724 Bytes
# ZeroGPU provides the CUDA torch build; do not pin torch hard.
# Modern set — paired with a recent Gradio (installed by HF from README `sdk_version`).
# Do NOT add a `gradio` line here; it conflicts with the Spaces runtime.
# The api_info "bool is not iterable" bug is handled by a monkeypatch in app.py.
spaces
torch
torchvision
diffusers>=0.31.0
transformers>=4.44.0
accelerate>=0.33.0
peft>=0.12.0
safetensors>=0.4.3
huggingface_hub>=0.25.0
sentencepiece
protobuf
Pillow
numpy
opencv-python-headless
# Face-identity mode (IP-Adapter FaceID). Heavy; comment out if you don't use Face mode.
insightface==0.7.3
onnxruntime
# Pose lock mode (ControlNet OpenPose) — pose detection from the uploaded image.
controlnet_aux