mirror of
https://github.com/comfyanonymous/ComfyUI.git
synced 2026-03-08 10:47:32 +08:00
9250191c65
DynamicVRAM's on-demand model loading/offloading conflicted with process isolation in three ways: RPC tensor transport stalls from mid-call GPU offload, race conditions between model lifecycle and active RPC operations, and false positive memory leak detection from changed finalizer patterns. - Marshal CUDA tensors to CPU before RPC transport for dynamic models - Add operation state tracking + quiescence waits at workflow boundaries - Distinguish proxy reference release from actual leaks in cleanup_models_gc - Fix init order: DynamicVRAM must initialize before isolation proxies - Add RPC timeouts to prevent indefinite hangs on model unavailability - Prevent proxy-of-proxy chains from DynamicVRAM model reload cycles - Add torch.device/torch.dtype serializers for new DynamicVRAM RPC paths - Guard isolation overhead so non-isolated workflows are unaffected - Migrate env var to PYISOLATE_CHILD
37 lines
491 B
Plaintext
37 lines
491 B
Plaintext
comfyui-frontend-package==1.39.19
|
|
comfyui-workflow-templates==0.9.5
|
|
comfyui-embedded-docs==0.4.3
|
|
torch
|
|
torchsde
|
|
torchvision
|
|
torchaudio
|
|
numpy>=1.25.0
|
|
einops
|
|
transformers>=4.50.3
|
|
tokenizers>=0.13.3
|
|
sentencepiece
|
|
safetensors>=0.4.2
|
|
aiohttp>=3.11.8
|
|
yarl>=1.18.0
|
|
pyyaml
|
|
Pillow
|
|
scipy
|
|
tqdm
|
|
psutil
|
|
alembic
|
|
SQLAlchemy
|
|
av>=14.2.0
|
|
comfy-kitchen>=0.2.7
|
|
comfy-aimdo>=0.2.4
|
|
requests
|
|
|
|
#non essential dependencies:
|
|
kornia>=0.7.1
|
|
spandrel
|
|
pydantic~=2.0
|
|
pydantic-settings~=2.0
|
|
PyOpenGL
|
|
glfw
|
|
|
|
pyisolate==0.9.1
|