lllyasviel / lllyasviel/stable-diffusion-webui-forge

Issue only with any type of Flux models

Open
#2,311 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
13k
Forks
1.7k
PR merge metrics
No merged PRs in 30d

Description

image
I´m trying to make images with Flux models: Unet, Checkpoints, trained, originals, and always have a warning. I always set 10000 Mb or less for GPU weights, leaving at least 2000 mb free for calculations (I have a Nvidia 3060 12Gb VRAM). Early versions of Forge for Flux worked fine, but now Forge always sees a few Mb free for calculations, no matter if I lower the GPU weigth value and the iteration gets stuck, at last only one image is made very slowly.
I tested Forge with Sd1.5 models and performs fine. It gets dizzy about free memory only with Flux

Contributor guide

No contributing guide indexed for this repository

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.

Research direction

No source files or tests are named. Start by reproducing generation with Flux and SD1.5 models on the reported Nvidia 3060 setup, then inspect how GPU memory available for calculations is detected and reported. Done means Flux generation no longer falsely reports insufficient memory or becomes stuck after one image.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.