A workbench for running and reviewing image-generation experiments. You can use it to:
- manage scenes and prompt parts
- combine and switch LoRAs
- run generation jobs on local or remote backends
- review results and validation images
- upscale images with
Real-ESRGAN(x2/x4)
Copy this prompt to your AI coding agent:
Deploy the zimage-eval-site project for me.
I will provide either:
1. a cloud server I already have, or
2. a local machine I want to run it on.
Your job is to handle setup, dependencies, configuration, startup, verification, and handoff.
Only ask me for the server access details or a necessary decision when you truly need it.
When you're done, give me:
- the app URL
- the API URL
- how to start it again
- how to stop it
If you want a more guided workflow, use the deployment playbook in docs/.