---
title: OuluDemo
app_file: app.py
sdk: gradio
sdk_version: 5.50.0
python_version: "3.10"
startup_duration_timeout: 1h
---

# Oulution Demo

This repo is configured for Hugging Face Spaces with `app.py` as the Gradio entrypoint.

## Deploy

1. Create a new Hugging Face Space with `Gradio` as the SDK.
2. Push this repository to the Space.
3. Add `DASHSCOPE_API_KEY` as a Space secret for the current `api_endpoint.json`.
4. Let the Space build. It will start from `app.py`.

If you switch providers later, update `api_endpoint.json` to match the schema in `docs/model_support.md` and add the matching secret, such as `OPENAI_API_KEY`.

## Optional Environment Variables

- `FASTCHAT_REGISTER_API_FILE`: alternate model config file, default `api_endpoint.json`
- `FASTCHAT_CONTROLLER_URL`: controller URL if you want worker-backed models
- `FASTCHAT_CONCURRENCY`: queue concurrency, default `10`
- `FASTCHAT_MODEL_LIST_REQUEST_TIMEOUT`: timeout in seconds for controller model-list calls, default `10`
- `FASTCHAT_VISION_ARENA`: set to `true` to show vision tabs
- `FASTCHAT_HIDE_ARENA_BATTLE`: set to `false` to show the battle tab

If `FASTCHAT_CONTROLLER_URL` is set, make sure it is reachable from the Hugging Face Space container. A controller on `localhost` or on a private machine will not be reachable from Spaces.

## Local Check

```bash
source /opt/anaconda3/etc/profile.d/conda.sh
conda activate oulu_demo
pip install -r requirements.txt
python app.py
```