zero_gpumodel / README.md
malek391's picture
Update README.md
0d2af9d verified
|
Raw
History Blame Contribute Delete
1.37 kB
---
title: My Custom Flutter Coder
emoji: πŸš€
colorFrom: blue
colorTo: purple
sdk: gradio
sdk_version: 5.9.0
app_file: app.py
pinned: false
hardware: zero-a10g
---
# My Custom Flutter Coder (ZeroGPU)
Free cloud hosting for `malek391/my-custom-flutter-coder` via [Hugging Face Spaces ZeroGPU](https://huggingface.co/docs/hub/spaces-zerogpu).
## Deploy (one time)
1. Create a new Space at [huggingface.co/new-space](https://huggingface.co/new-space)
2. Choose **Gradio** SDK
3. Upload `app.py`, `requirements.txt`, and this `README.md`
4. In Space **Settings β†’ Hardware**, select **ZeroGPU** (`zero-a10g`)
5. Add a Space secret `HF_TOKEN` with your Hugging Face access token (read access to the model)
6. Wait for the build to finish β€” your Space URL will be like `https://YOUR-USER-my-custom-flutter-coder.hf.space`
## Connect Flutter IDE
In the IDE **Settings β†’ Hugging Face Space URL**, paste:
```
https://YOUR-USER-my-custom-flutter-coder.hf.space
```
Or in `.env`:
```
HUGGINGFACE_SPACE_URL=https://YOUR-USER-my-custom-flutter-coder.hf.space
HUGGINGFACE_ACCESS_TOKEN=hf_...
```
## Limits (free tier)
- ~5 minutes GPU time per day (shared queue)
- Cold starts when the Space sleeps
- Best for development β€” not 24/7 production
## API
The IDE calls the Gradio endpoint `POST /gradio_api/call/chat` with:
`[system_prompt, user_prompt, temperature]`