api-inference-community icon indicating copy to clipboard operation
api-inference-community copied to clipboard

Use Lora with hosted inference API

Open lionel-alves opened this issue 2 years ago • 2 comments

Hi,

I would like to choose a model and a lora and generate images with it using the hosted inference API As showed here: https://huggingface.co/pcuenq/pokemon-lora

If parameters has a lora_model_id use merge with the lora model.

lionel-alves avatar May 30 '23 13:05 lionel-alves

If you use a diffusers LoRA I think this is already what will happen, the lora will download the mother model and use them.

@pcuenca that's correct, right ?

Narsil avatar Jun 05 '23 14:06 Narsil

If you use a diffusers LoRA I think this is already what will happen, the lora will download the mother model and use them.

Yes, exactly! The mother model is referenced in the base_model metadata field as shown here: https://huggingface.co/pcuenq/pokemon-lora/blob/main/README.md?code=true#L4

pcuenca avatar Jun 05 '23 15:06 pcuenca