Spaces:
Running
on
Zero
Error with Flux Models
Whenever I try to use a Flux model, I get different errors depending on the model. None of them seem to work. The most common error is: FluxPipeline.init() missing 1 required positional argument: 'scheduler'
Pony and Illustrious models work fine.
Is there any solution to this?
Try this.
https://huggingface.co/spaces/John6666/flux-lora-the-explorer
It just seems to stall and doesn't generate an image.
I'm not sure what the cause is...🤔
How about the original DiffuseCraft?
https://huggingface.co/spaces/r3gm/DiffuseCraft
I'm not sure what the cause is...🤔
How about the original DiffuseCraft?
https://huggingface.co/spaces/r3gm/DiffuseCraft
Yes, multiple Flux models are working with that one.
Yes, multiple Flux models are working with that one.
That's a big hint. I tried changing some things, but it seems to have fixed it when I fixed the Diffusers version to an older one. Anyway, thank you!
Yes, multiple Flux models are working with that one.
That's a big hint. I tried changing some things, but it seems to have fixed it when I fixed the Diffusers version to an older one. Anyway, thank you!
It still gives me the same error, sadly.
I thought it worked for a moment yesterday, but maybe I was imagining it.
Hmm, stablepy is throwing up a mysterious error. I think it's probably because of some library or Spaces change since last December, but no matter what I do, I can't fix it...
At the moment, it seems impossible for me to fix it.😭
I thought it worked for a moment yesterday, but maybe I was imagining it.
Hmm, stablepy is throwing up a mysterious error. I think it's probably because of some library or Spaces change since last December, but no matter what I do, I can't fix it...
At the moment, it seems impossible for me to fix it.😭
🤔
Weird that it still works on the original DiffuseCraft.
Weird that it still works on the original DiffuseCraft.
True.
I found a strange part in the code, so I tried to fix it. It seems to have been introduced during the update process.
It feels like it's improved.🤔
That seems to have fixed it! Thank you!