Elvenson/diffuser_inference

Stable Diffusion v1-5 Custom Inference

This repo is for running diffusion custom inference endpoints with prompts and an optional image as inputs (Unlike normal text-to-image inference). To achieve this goal, this repo implements a handler.py script. For more information regarding custom inference, please visit this link. For more information about the model, license and limitations please check the original model card or diffusion documentation.

Local test custom handler

To test custom inference locally, please run the following command:

 python local_request.py --prompts="whale in the universe" --image="test_image.jpg"

Note: --image parameter is optional.