plano/model_server
Adil Hafeez b76f3b84ee
use uv
2025-10-22 14:59:57 -07:00
..
.vscode better model names (#517) 2025-07-11 16:42:16 -07:00
src swap out python to sys.executable to support uv tool install workflows (#544) 2025-07-28 00:48:35 -07:00
tests Integrate Arch-Function-Chat (#449) 2025-04-15 14:39:12 -07:00
__init__.py model server build (#127) 2024-10-06 18:21:43 -07:00
Dockerfile Use intent model from archfc to pick prompt gateway (#328) 2024-12-20 13:25:01 -08:00
Dockerfile.gpu Use intent model from archfc to pick prompt gateway (#328) 2024-12-20 13:25:01 -08:00
poetry.lock release 0.3.14 (#577) 2025-09-29 23:11:43 -07:00
pyproject.toml use uv 2025-10-22 14:59:57 -07:00
README.md use uv 2025-10-22 14:59:57 -07:00
uv.lock use uv 2025-10-22 14:59:57 -07:00

Model Server Package

This model server package is a dependency of the Arch intelligent prompt gateway. It should not be used alone. Please refer to the quickstart-guide for more details on how to get start with Arch.

Local development

You can start/stop the local server via the CLI entry point exposed by this package.

Using uv (recommended):

uv run model_server --help
# run in foreground (stays attached until Ctrl+C)
uv run model_server start --port 51000 --foreground
# run in background (then stop using the CLI)
uv run model_server start --port 51000
uv run model_server stop

Alternative without uv:

python -m src.cli --help
# foreground
python -m src.cli start --port 51000 --foreground
# background
python -m src.cli start --port 51000
python -m src.cli stop

The FastAPI app lives at src.main:app and exposes a health check at /healthz.