Support for Diffusers?

#5
by tintwotin - opened

No support for Diffusers?

Stability AI org

Coming in hours.

Or minutes? :D

omar being spicy. i love it

Entry Not Found for url: https://huggingface.co/stabilityai/stable-diffusion-3-medium/resolve/main/model_index.json.

For anyone else as impatient as myself - you can download pull requests as "revisions", for example the below pulls from the PR with diffusers weights and starts downloading the model:

pipe = StableDiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-3-medium", revision="refs/pr/26")

Not sure if this will work, but let's see ;)

Will this model work well with the StableDiffusionInpaintPipeline?

Any updates guys?

huoju changed discussion status to closed

Still getting this issue: cannot import name 'StableDiffusion3Pipeline' from 'diffusers' even though I did update diffusers.

A diffusers release is incoming!

This comment has been hidden

The unet config is still missing in the repo

probably because it's....a..... wait for it.... Transformer! not a u-net.

ahh.... wait for it..... Thanks a lot for the clarification 🤗.

If you could explain how can I fine tune it because following this guide https://huggingface.co/docs/diffusers/v0.30.0/training/text2image fails with unet config missing would be great help.

using a better tool most likely - those are just examples meant for developers to help understand how the library works.

here is a quickstart guide: https://github.com/bghira/SimpleTuner/blob/main/documentation/quickstart/SD3.md

Sign up or log in to comment