NVIDIA / NVIDIA/TensorRT

stable diffusion quantization in inpainting task is poor

Open
#4,212 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Module:Demo triaged
Dominant language
C++
Stars
13.4k
Forks
2.4k
Avg merge
5d 3h
Merged PRs (30d)
2

Description

i have completed stable diffusion quantization in txt2img as demo shows.
the result is very good.
when i want to transfer sd quantization in inpainting task, i meet the problem that the quantization result is very poor.
the mask area is still noise and the quantized unet model can not inpaint any thing.
Do you konw the problem and anyone can help me?

Contributor guide

Open the contributing guide

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

Use the existing txt2img quantization demo as the baseline, then reproduce the reported behavior with the inpainting task and quantized UNet. The issue names no files, tests, or runnable entry point, so first identify the relevant Stable Diffusion inpainting and quantization paths. Done means the mask area is no longer noise and the quantized model can inpaint.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning
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.