https://github.com/mcaimi/text-to-image-frontend
Frontend application for the Text-To-Image OCP AI Demo
https://github.com/mcaimi/text-to-image-frontend
Last synced: 3 months ago
JSON representation
Frontend application for the Text-To-Image OCP AI Demo
- Host: GitHub
- URL: https://github.com/mcaimi/text-to-image-frontend
- Owner: mcaimi
- License: gpl-3.0
- Created: 2024-06-27T15:27:44.000Z (about 2 years ago)
- Default Branch: main
- Last Pushed: 2024-12-02T11:43:58.000Z (over 1 year ago)
- Last Synced: 2025-01-16T21:16:02.739Z (over 1 year ago)
- Language: Python
- Size: 1.13 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Text To Image Demo Frontend
This is the frontend application for the [Text To Image Stable Diffusion Demo](https://github.com/mcaimi/text-to-image-demo)
It exposes a way to perform inference with a Stable Diffusion compatible model in the backend.
It is built with FastAPI and Gradio and works both locally (in a virtual environment for example) or inside Openshift.
It has however a dependency on the KServe model server engine that runs in OCP AI. Look [here](https://github.com/mcaimi/kserve-diffusers-demo) for the backend code.
## Run Locally
Normally the software will run in "remote" mode, meaning that it will use the "INFER_URL" environment variable to locate the API endpoint to use to perform inference.
Usually that endpoint is hosted remotely on Openshift AI.
The application expects to find model checkpoint files in the `models/stable-diffusion` folder.
```bash
# development mode
$ fastapi dev
# production mode
$ fastapi run
```
Run on Openshift (assuming namespace stable-diffusion and model server instance stable-diffusion):
```bash
$ oc new-app --name stable-diffusion-frontend -e INFER_URL=$(oc get route -n istio-system stable-diffusion-stable-diffusion -o jsonpath='{.spec.host}') quay.io/marcocaimi/stable-diffusion-frontend:latest
```