Skip to content

Throttle dynamic VRAM prepare logging #13704

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
kijai wants to merge 4 commits into
base: master
Choose a base branch
from
+6 −1
Open

Throttle dynamic VRAM prepare logging #13704

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
f19fb21
Throttle dynamic VRAM load logging
kijai May 4, 2026
778e5e8
Throttle only inside when tqdm running
kijai May 4, 2026
1eb072f
Merge remote-tracking branch 'upstream/master' into dynvram_log
kijai May 5, 2026
fb1c422
Merge branch 'master' into dynvram_log
kijai May 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion comfy/model_patcher.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1651,7 +1651,10 @@ def force_load_param(self, param_key, device_to):
self.model.model_loaded_weight_memory += casted_buf.numel() * casted_buf.element_size()

force_load_stat = f" Force pre-loaded {len(self.backup)} weights: {self.model.model_loaded_weight_memory // 1024} KB." if len(self.backup) > 0 else ""
logging.info(f"Model {self.model.__class__.__name__} prepared for dynamic VRAM loading. {allocated_size // (1024 ** 2)}MB Staged. {num_patches} patches attached.{force_load_stat}")
log_key = (self.patches_uuid, allocated_size, num_patches, len(self.backup), self.model.model_loaded_weight_memory)
Comment thread
kijai marked this conversation as resolved.
level = logging.DEBUG if getattr(self, "_last_prepare_log_key", None) == log_key else logging.INFO
self._last_prepare_log_key = log_key
logging.log(level, f"Model {self.model.__class__.__name__} prepared for dynamic VRAM loading. {allocated_size // (1024 ** 2)}MB Staged. {num_patches} patches attached.{force_load_stat}")

self.model.device = device_to
self.model.current_weight_patches_uuid = self.patches_uuid
Expand Down
Loading