Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.quanux.org/llms.txt

Use this file to discover all available pages before exploring further.

This guide walks you through the fastest path from zero to a running QuanuX environment. By the end, you’ll have the backend server and Tauri research cockpit running locally, your secrets stored securely, and a backtest executing against historical data. The steps below assume you’re on Linux or macOS and have the prerequisites installed — see the Installation guide if you need to set those up first.
1

Install QuanuX

The fastest way to install QuanuX is through the official Conda channel. This installs the platform and all Python dependencies into a managed environment with no dependency conflicts.
conda config --add channels conda-forge
conda config --add channels QuanuX
conda install quanux
If you prefer pip or want to install from source, see the full Installation guide for all three install paths.
QuanuX requires Python 3.10+, Node.js v20+, and pnpm. The Conda install handles Python dependencies automatically, but you’ll need Node.js and pnpm available separately for the cockpit.
2

Configure your secrets

QuanuX stores broker API keys, infrastructure credentials, and other secrets in your operating system’s keyring — nothing is written to disk in plaintext. Run the interactive setup CLI to configure them:
python -m server.cli.setup_secrets
The CLI will prompt you for each credential. At minimum, configure the keys for the brokerage you intend to connect to. You can add more credentials later at any time by re-running the wizard or using quanuxctl <integration> <field> "<value>".
See Managing secrets for the full list of supported integrations and how to set credentials for Rithmic, IBKR, TopstepX, and Tradovate.
3

Start the core server

In a dedicated terminal, start the QuanuX backend server with uvicorn:
uvicorn server.app.main:app --host 0.0.0.0 --port 8080 --reload
The --reload flag enables hot-reloading during development. For production nodes, omit it. The server exposes the QuanuX API on port 8080 and manages connections to your configured brokerages and data sources.
4

Launch the research cockpit

In a second terminal, start the Tauri desktop cockpit. This is your primary interface for strategy development, backtesting, and observing live execution:
pnpm -C client/desktop/tauri-app tauri dev
The cockpit window will open automatically. On first launch, it connects to the server running on localhost:8080.
You can also run the cockpit as a web app without the Tauri desktop wrapper. Use pnpm run web from the project root to launch the browser-based version instead.
5

Run your first backtest

With the server running and the cockpit open, navigate to the Strategy Forge section in the cockpit sidebar. From there you can:
  1. Create a new strategy using the Python editor or the AI Strategy Architect.
  2. Select a historical dataset (Databento L3 data is the primary source).
  3. Run the strategy through the Crucible backtesting engine.
The Crucible engine uses the same code paths as live execution, so backtest results have binary replay parity with live structural executions — no drift between research and production.For a detailed walkthrough of writing and backtesting strategies, see Writing strategies and Backtesting.

​
What’s next

Core concepts

Understand how QuanuX nodes are structured and how the execution plane works.

Strategy Forge

Learn how to develop, compile, and deploy strategies through the Forge.

Brokerages

Connect to your brokerage or prop firm through Rithmic, TopstepX, or IBKR.

CLI Reference

Use quanuxctl to manage nodes, deploy engines, and monitor your cluster.