Closed aresnow1 closed 7 months ago
@Bojun-Feng Hi, any interest? Or any suggestions we can discuss. :)
@aresnow1 Sounds like an interesting feature! However, I am worried about locally testing the UI on my computer, as my Mac laptop has no GPU support and has only 8GB of RAM. If there are any small text2image or image2image models I can run on my laptop, I would be more than happy to implement the UI for it!
As for suggestions, we can probably look at this open source repo.
I envision the UI to look like this Hugging Face space, except with bigger input text box.
Hi, could you try this one(https://huggingface.co/stabilityai/sd-turbo ) on your laptop?
I will create a PR to add sd turbo support.
As we support stable diffusion model for text2image, a simple UI is needed to let users generate images on web pages. Like chat UI, we can launch a gradio page for a quick implementation.