Can’t access custom public LoRa models

#1
by fffiloni - opened

the app cannot access custom public LoRa models, may i suggest to tweak some fine-grained option to the private HF token, so it can access public models repos ?

do you have an example of a public LoRA that does not work?

Sure for example: https://huggingface.co./fffiloni/deep-blue-v2

Also i tried with listed Flux LoRAs from there: https://huggingface.co./models?inference_provider=fal-ai&pipeline_tag=text-to-image&sort=trending

—> loading the model fails.

I tried to duplicate the space to reproduce and PR a solution, but couldn’t duplicate the space (got a build error: Error while cloning repository), see: https://huggingface.co./spaces/fffiloni/Blazingly-fast-LoRA-perso

@julien-c @reach-vb do the custom model should be referenced on FAL too ?

@julien-c @reach-vb do the custom model should be referenced on FAL too ?

no, no need

note that the inference widget on https://huggingface.co./fffiloni/deep-blue-v2 does work (though it does not seem to really pick up the trigger word/style? not sure) and it's powered by the Fal integration

Your need to confirm your account before you can post a new comment.

Sign up or log in to comment