ai-toolkit icon indicating copy to clipboard operation
ai-toolkit copied to clipboard

Modal instructions not working - Huggingface problem

Open faermi opened this issue 11 months ago • 2 comments

This is for bugs only

Did you already ask in the discord?

Yes

You verified that this is a bug and not a feature request or question by asking in the discord?

Yes

Describe the bug

I am having a problem with the Modal approach. I have succesfully trained a LoRa with RunPod, but with Modal I can not get the Flux model from huggingface. I have correctly configured huggingface-cli and also copied a .env with the token in /ai-toolkit and /ai-toolkit/config just in case... but nothing. I keep on getting:

" black-forest-labs/FLUX.1-dev is not a local folder and is not a valid model identifier listed on │ │ 'https://huggingface.co/models' │ │ If this is a private repository, make sure to pass a token having permission to this repo with token or log in │ │ with huggingface-cli login. "

As I said, I have had no problem using RunPod before, but Modal just refuses to work. I am using Debian bookworm in WSL. As I indicated, I have had no problems training LoRas with RunPod, but this keeps happening with Modal.

Image

faermi avatar Feb 06 '25 11:02 faermi

Having the same issue.

vivek-kumar-poddar avatar Feb 16 '25 08:02 vivek-kumar-poddar

Has anyone been able to run it correctly in Modal? How?

faermi avatar Feb 26 '25 09:02 faermi