Dhruv Nair

@gulatis A LoRA isn't the same thing as a textual inversion. So it makes sense that the loading would fail.

Yeah makes sense to add single file support for `T2IAdapter`. @JemiloII Do you have an example hosted checkpoint we could use to test? And would you be interested in adding...

Hi @vladmandic could you please provide a code example I could use to reproduce/test the issue you're facing?

Hi @dferguson992 You could zip the entire cache dir and unzip on your endpoint machine/script. If you're using the HF Cache mechanism, and the models have been saved to the...

Yes we can 👍🏽 Was planning on adding the support this week, but definitely a good one for the community as well.

Hi @vladmandic we merged #8819 with single file support for MotionAdapter a little while back.

Closing since this has been completed.

@saeedkhanehgir Closing this issue for now since the pipeline isn't throwing an error. For help with dealing with long prompts, it might be better to open a thread in the...

Actually this test and functionality should be removed. We don't support configuring models via dict in `from_pretrained` so `from_single_file` shouldn't support it either. `config` should be a model repo id...

Sorry I missed this. Left a comment on a spelling mistake and I believe @yiyixuxu's comment still needs to be addressed. https://github.com/huggingface/diffusers/pull/7930/files#r1618014761 Good to merge once conflicts are resolved and...