wavymulder

/

collage-diffusion

like 21

Alright, I've tried at least 5 different times... I am unable to convert this to Diffusers using any Huggingface space I can find.

hi @wavymulder it seems like the SD to diffusers its not working correcly,

you might need to follow the steps for the conversion here https://discuss.huggingface.co/t/how-to-convert-ckpt-to-diffusers-format/35635

it seems like the original config file is missing on your repo
--original_config_file - the original config file. This should be a file that ends with a yaml suffix. In case no file is present you should try to find the corresponding file in the github repo the .ckpt model originated from. Some examples:

It's worked every other time for every other model using the same method, kinda at a loss as to what I need to do fix this. The Space allows to select for model version, which should select the v.1 inference yaml.

great point, can you jump to that thread and tag @patrickvonplaten ?

could you please @wavymulder try again? https://huggingface.co/spaces/diffusers/sd-to-diffusers I've reboot the space , is your origiinal model v1? or v2?

@radames thank you, the original model is v1. I will retry in a few moments, am currently downloading the ckpt to ensure it uploaded correctly and is not corrupted or something.