Skip to content

Installation

This page is for anyone setting up QuOptuna for the first time. By the end you’ll have the package installed and the app running at http://localhost:8000.

RequirementDetails
Python3.11 or 3.12
RAM4 GB minimum, 8 GB recommended for quantum models
InternetNeeded for UCI datasets and LLM-generated reports
Node.js18+ — only if you run the frontend in dev mode

With uv:

Terminal window
uv pip install quoptuna

Or with pip:

Terminal window
pip install quoptuna

For an editable development install with extra tooling:

Terminal window
uv pip install -e ".[dev]"

QuOptuna can run three ways. Most users want packaged mode.

ModeCommandPortsWhen to use
Packageduvx quoptuna or uv run quoptuna run:8000 (UI + API)Everyday use; simplest, one process
Devmake dev:8000 API + :3000 UIContributing with hot reload
Dockerdocker compose up --build:3000 frontend + :8000 backendIsolated, reproducible environment

Serves the web UI and JSON API on a single port. It opens your browser automatically and prints a gradient ASCII banner.

Terminal window
uvx quoptuna

Running uvx quoptuna (or uv run quoptuna) with no subcommand is the same as quoptuna run. Useful options:

Terminal window
uv run quoptuna run --port 8001 --no-browser
uv run quoptuna run --host 0.0.0.0
uv run quoptuna run --streamlit # launch the legacy Streamlit dashboard

Runs FastAPI and the Next.js frontend separately so both hot-reload. This needs Node 18+ for the frontend.

Terminal window
make dev # FastAPI on :8000 + Next.js on :3000

Individual targets are also available: make run_backend, make run_frontend, make run_streamlit, and make run_cli.

Terminal window
docker compose up --build

The docker-compose.yml defines the frontend on :3000 and the backend on :8000.

  1. Start the app: uvx quoptuna.
  2. Open the web UI at http://localhost:8000 — you should see the 6-step wizard.
  3. Open the interactive API docs at http://localhost:8000/api/docs to confirm the API is live.

The JSON API lives under http://localhost:8000/api/v1/....