Update README.md
Browse files
README.md
CHANGED
|
@@ -184,18 +184,6 @@ docker-compose up --build
|
|
| 184 |
### Continuous Integration
|
| 185 |
This repository includes a GitHub Actions CI to lint, type-check, and run tests on PRs/commits to `main`.
|
| 186 |
|
| 187 |
-
### Hugging Face Spaces (private)
|
| 188 |
-
You can deploy the Gradio app privately:
|
| 189 |
-
1) In Space settings, set `Visibility: Private`.
|
| 190 |
-
2) Use SDK: Gradio; Hardware: pick CPU or T4 (GPU recommended).
|
| 191 |
-
3) Add Secrets/Variables (Settings → Variables & secrets):
|
| 192 |
-
- `MODEL_PATH`: path or HF repo id (e.g., `ainowmk/MK-LLM-Mistral`)
|
| 193 |
-
- `API_URL` (optional): if pointing to external FastAPI
|
| 194 |
-
4) Set `app_file` to `inference/gradio_app.py`.
|
| 195 |
-
5) If using local generation in Space, ensure the model repo is accessible and quantization flags are suitable.
|
| 196 |
-
|
| 197 |
-
Update existing Space: push the updated repo files; Spaces will rebuild automatically.
|
| 198 |
-
|
| 199 |
### Constraints (reproducible installs)
|
| 200 |
To install with pinned versions:
|
| 201 |
```bash
|
|
|
|
| 184 |
### Continuous Integration
|
| 185 |
This repository includes a GitHub Actions CI to lint, type-check, and run tests on PRs/commits to `main`.
|
| 186 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 187 |
### Constraints (reproducible installs)
|
| 188 |
To install with pinned versions:
|
| 189 |
```bash
|