xinyun/ComfyUI
1
0
Fork 0
mirror of https://git.datalinker.icu/comfyanonymous/ComfyUI synced 2025-12-10 14:34:25 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix on-load VRAM OOM (#11144)

Browse Source 
slow down the CPU on model load to not run ahead. This fixes a VRAM on
flux 2 load.

I went to try and debug this with the memory trace pickles, which needs
--disable-cuda-malloc which made the bug go away. So I tried this
synchronize and it worked.

The has some very complex interactions with the cuda malloc async and
I dont have solid theory on this one yet.

Still debugging but this gets us over the OOM for the moment.
This commit is contained in:
rattus 2025-12-07 09:42:09 +10:00 committed by GitHub
parent 50ca97e776
commit 4086acf3c2
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
comfy/model_patcher.py
View File

@ -762,6 +762,8 @@ class ModelPatcher:
key = "{}.{}".format(n, param) key = "{}.{}".format(n, param)
self.unpin_weight(key) self.unpin_weight(key)
self.patch_weight_to_device(key, device_to=device_to) self.patch_weight_to_device(key, device_to=device_to)
if comfy.model_management.is_device_cuda(device_to):
torch.cuda.synchronize()
logging.debug("lowvram: loaded module regularly {} {}".format(n, m)) logging.debug("lowvram: loaded module regularly {} {}".format(n, m))
m.comfy_patched_weights = True m.comfy_patched_weights = True