Skip to content

[enhancement]: Support TAESD to display better progress image #4056

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

Jump to bottom
Open
1 task done
liuchibing opened this issue Jul 29, 2023 · 3 comments · Fixed by #4316
Open
1 task done

[enhancement]: Support TAESD to display better progress image #4056

liuchibing opened this issue Jul 29, 2023 · 3 comments · Fixed by #4316
Labels
enhancement New feature or request good first issue Good for new contributors

Comments

@liuchibing
Copy link

Is there an existing issue for this?

  • I have searched the existing issues

Contact Details

No response

What should this feature add?

Maybe add a switch in settings for users to choose TAESD when displaying progress images, instead of current mathematical method.

TAESD is very tiny autoencoder which uses the same "latent API" as Stable Diffusion's VAE*. TAESD can decode Stable Diffusion's latents into full-size images at (nearly) zero cost.

Here is the repo: https://github.com/madebyollin/taesd

Alternatives

No response

Aditional Content

No response
@liuchibing liuchibing added the enhancement New feature or request label Jul 29, 2023
@psychedelicious psychedelicious added the good first issue Good for new contributors label Jul 31, 2023
@psychedelicious
Copy link
Collaborator

Looks like diffusers may support this natively soon: huggingface/diffusers#4384

@keturn
Copy link
Contributor

keturn commented Sep 1, 2023

Initial TAESD integration in #4316

@keturn keturn linked a pull request Sep 1, 2023 that will close this issue
feature: support TAESD - Tiny Autoencoder for Stable Diffusion #4316
Merged
6 tasks
@keturn
Copy link
Contributor

keturn commented Sep 20, 2023

that PR added support for using TAESD as a VAE, but we've yet to integrate it for progress images, so this is still open.

@keturn keturn reopened this Sep 20, 2023
@zealsprince zealsprince mentioned this issue Sep 28, 2023
Closed
1 task
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request good first issue Good for new contributors
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

feature: support TAESD - Tiny Autoencoder for Stable Diffusion invoke-ai/InvokeAI
3 participants
@keturn @psychedelicious @liuchibing