Update Gunicorn startup instructions in API documentation
This commit is contained in:
@@ -102,8 +102,11 @@ light-server --llm-binding ollama --embedding-binding ollama
|
|||||||
For production deployments, it's recommended to use Gunicorn as the WSGI server to handle concurrent requests efficiently. LightRAG provides a dedicated Gunicorn startup script that handles shared data initialization, process management, and other critical functionalities.
|
For production deployments, it's recommended to use Gunicorn as the WSGI server to handle concurrent requests efficiently. LightRAG provides a dedicated Gunicorn startup script that handles shared data initialization, process management, and other critical functionalities.
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
# Start with run_with_gunicorn.py
|
# Start with lightrag-gunicorn command
|
||||||
python run_with_gunicorn.py --workers 4
|
lightrag-gunicorn --workers 4
|
||||||
|
|
||||||
|
# Alternatively, you can use the module directly
|
||||||
|
python -m lightrag.api.run_with_gunicorn --workers 4
|
||||||
```
|
```
|
||||||
|
|
||||||
The `--workers` parameter is crucial for performance:
|
The `--workers` parameter is crucial for performance:
|
||||||
|
Reference in New Issue
Block a user