[GH-ISSUE #2648] enh: foocus image generation engine support #12962

Closed
opened 2026-04-19 19:46:34 -05:00 by GiteaMirror · 4 comments
Owner

Originally created by @tuantupharma on GitHub (May 29, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/2648

I hope open-webui support Fooocus in mage Generation Engine. It light weight and the image prompt is sometime better than comfyui, automatic1111.

Originally created by @tuantupharma on GitHub (May 29, 2024). Original GitHub issue: https://github.com/open-webui/open-webui/issues/2648 I hope open-webui support Fooocus in mage Generation Engine. It light weight and the image prompt is sometime better than comfyui, automatic1111.
GiteaMirror added the good first issuenon-core labels 2026-04-19 19:46:35 -05:00
Author
Owner

@spammenotinoz commented on GitHub (Jun 6, 2024):

FYI: If it helps in the current version, I use the models feature to create a custom model called "Stable Diffusion" with a system prompt.
"You are an expert in providing a short and concise image description for Stable Diffusion. Provide no other explanation or reasoning, just a single descriptive prompt suitable for image generation."
Not claiming it's the best system prompt, but is working well. For the model I am using a rather uncensored model from openrouter. "openrouter/neversleep/llama-3-lumimaid-8b"
You also have the EDIT icon, and clicking the "Generate Image" icon again and again creates more images.
What may work well is a dummy model called "IMAGE" that doesn't use any model, just sends prompts directly to the back-end image api.

<!-- gh-comment-id:2151277817 --> @spammenotinoz commented on GitHub (Jun 6, 2024): FYI: If it helps in the current version, I use the models feature to create a custom model called "Stable Diffusion" with a system prompt. "You are an expert in providing a short and concise image description for Stable Diffusion. Provide no other explanation or reasoning, just a single descriptive prompt suitable for image generation." Not claiming it's the best system prompt, but is working well. For the model I am using a rather uncensored model from openrouter. "openrouter/neversleep/llama-3-lumimaid-8b" You also have the EDIT icon, and clicking the "Generate Image" icon again and again creates more images. What may work well is a dummy model called "IMAGE" that doesn't use any model, just sends prompts directly to the back-end image api.
Author
Owner

@ItsNoted commented on GitHub (Jun 7, 2024):

Came here to see if Fooocus could be supported. This project has been fantastic for image generation and would be awesome if we could implement this into open webui.

<!-- gh-comment-id:2154711082 --> @ItsNoted commented on GitHub (Jun 7, 2024): Came here to see if Fooocus could be supported. This project has been fantastic for image generation and would be awesome if we could implement this into open webui.
Author
Owner

@jmtemmos commented on GitHub (Jun 28, 2024):

Hello. It would be great to have Fooocus support :) agreed :)

<!-- gh-comment-id:2197622744 --> @jmtemmos commented on GitHub (Jun 28, 2024): Hello. It would be great to have Fooocus support :) agreed :)
Author
Owner

@dot-Justin commented on GitHub (Oct 28, 2024):

@tjbck I see you've self-assigned this, this repo may be useful: https://github.com/mrhan1993/Fooocus-API

<!-- gh-comment-id:2440333532 --> @dot-Justin commented on GitHub (Oct 28, 2024): @tjbck I see you've self-assigned this, this repo may be useful: https://github.com/mrhan1993/Fooocus-API
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#12962