Server commands¶
Cmd: uni server up¶
This command builds and starts the UNI server.
| Parameter | Description |
|---|---|
--attach |
Attach current terminal to process |
--port <n> |
Port number (default: 6001) |
--dev |
Run in development mode with hot reload |
--debug |
Enable debugpy on port 5678 |
--new-venv |
Reinstall all packages (for troubleshooting) |
Shorthand
The command uni <args> is short for uni server up --attach <args>.
Auto-update¶
To check for updates, add the --pull flag:
| Parameter | Description |
|---|---|
--pull |
Pull latest release before starting |
--pull latest |
Pull latest development version before starting |
--pull @v1.0.0 |
Check out a specific git tag |
--pull #abc1234 |
Check out a specific commit |
Options for llama.cpp¶
Arguments prefixed --llama: are passed through to the llama.cpp server:
uni server up --llama:main-gpu 1 --llama:mlock
Refer to the llama.cpp server docs for available options.
Cmd: uni server logs¶
This command streams the UNI server logs to your current terminal window.
Press Ctrl+C to exit.
Logs are also saved to a file. Check the <uni-install-dir>/user/logs directory.
Cmd: uni server down¶
This command shuts down the UNI server and cleans up any resources.
Your settings and data stay saved.