Verified project record

OpenByteInc/QuantDinger

QuantDinger is a self-hosted application that turns trading ideas into tested Python strategies, covering backtesting, paper trading, live execution, and monitoring in one stack. It supports trading across crypto, stock, and forex brokers.

Project overview

It addresses a documented gap for independent traders and small teams who lack a self-hosted stack that combines strategy authoring, server-side backtesting with robustness validation, and safe multi-asset execution, while keeping strategy code, data, and credentials under the operator's control.

Project type: MCP · AI Agent
Use cases: Coding & Development · Automation
Deployment: Refer to project documentation
License: Apache-2.0

Best for

Independent traders and small teams who want a self-hosted stack for authoring, testing, and executing their own Python trading strategies.

Key capabilities

Python Strategy API V2 with server-side backtesting, parameter optimization (random, grid, TPE), walk-forward validation, blind holdout, and robustness tests such as Deflated Sharpe and Monte Carlo.

Limitations and risks

QuantDinger can submit real orders when live trading is explicitly enabled; the project does not provide investment advice.
Manual Docker deployments retain the quantdinger/123456 default credentials, which are not suitable for internet-facing deployment; change them before any exposed use.

Getting started

A one-command installer downloads the GHCR Compose stack and starts it; Node.js and local Python are not required.

Evidence and sources

README: Turn trading ideas into Python strategies, backtests, paper trading, live execution, and monitoring — all in one self-hosted stack.
README: Prerequisites: Docker with Compose v2. Node.js and a local Python environment are not required.
README: Its local-first, self-hosted design keeps market data, strategy code, broker credentials, and deployment under the operator's control.
README: **Strategy evolution** searches declared tunable parameters with random, grid, or TPE optimization, using bar-count walk-forward validation and a final blind holdout.
README: the production overlay runs backend processes as a non-root user with a read-only root filesystem, dropped capabilities, and resource limits;

Source updated at: 2026-10-23T18:29:21Z

Open original GitHub project