Endpoint interface for inpainting models that require two images.

[OrderAndCh4oS]

Is your feature request related to a problem? Please describe.
There doesn't appear to be an endpoint interface for Stable Diffusion inpainting models that require two image files, the base image and a mask.

Describe the solution you'd like
It would be handy to have an interface for these models so that the Hosted inference API widget would work on the model card views.

Describe alternatives you've considered
I recently had to create an endpoint for ControlNet inpainting: OrderAndChaos/controlnet-inpaint-endpoint. This was based on the philschmid/stable-diffusion-2-inpainting-endpoint endpoint.

[Wauplin]

Hi @OrderAndCh4oS , I think this issue should be addressed in https://github.com/huggingface/api-inference-community. If I understand correctly, you'd like to have a type of task supported right? More or less make https://huggingface.co/philschmid/stable-diffusion-2-inpainting-endpoint an "official" task? I think this should make sense only if there is a huge demand for it. Custom endpoints in Inference Endpoints are precisely made to handle more complex queries such as this one.

If you are looking for a way to demo your model, the best is to build a Gradio app and deploy it as a Space. Your model and your Space will automatically be linked so that users can test it.

[OrderAndCh4oS]

@Wauplin, thanks for the response. Totally fair. I believe this may be quite a common requirement for many stable diffusion models, so thought it worth asking anyway. I've moved the question to: huggingface/api-inference-community#282.