There's a major discrepancy in RAM utilization with ComfyUI v43 and up.

Open
#17,648 1 comment 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
43/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Active
Tech stack
python, pytorch

Research direction

Start by comparing ComfyUI v42.1 with v43.0 and v44.0 while running the same Krea 2 generation on a 32 GB system, and inspect memory usage in Task Manager. The report does not identify a file or test, so first determine which change could explain the difference between process RAM and total system RAM. Done means reproducing the discrepancy and identifying its cause.

Written by the indexing model from the issue text.

Description

Potential Bug
Custom Node Testing
Expected Behavior

Generations (with Krea 2 models) should be using around 21 GB of RAM and total system RAM usage caps at 90%.

Actual Behavior

Generations (with Krea 2 models) caps at around 7 GB of RAM and total system RAM usage is uncapped, goes up to 99% for no reason.

Steps to Reproduce

Update ComfyUI to version 43.0 or 44.0 and try to do anything with it.

Debug Logs
N/A
Other

So I have 32GB of RAM. With ComfyUI v42.1, my generations would normally use about 21 GB of that according to task manager. And my total RAM utilization would hover somewhere between 75-90%. Yet on ComfyUI v43 and v44, my generations are only using about 7-12 GB of RAM, while total RAM utilization skyrockets to 90-99% and my system damn near crashes. So despite v43+ using less RAM according to task manager, it's somehow using more RAM that's unaccounted for. And now once again I find myself staying on a lower version of some software because the newer versions broke something. But this time its in the background rather than the atypical UI enshitification. What did you guys do??

Dominant language
Python
Stars
146k
Forks
17.4k
Avg merge
19h 44m
Merged PRs (30d)
178

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from Comfy-Org/ComfyUI

All issues in Comfy-Org/ComfyUI

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.