QuantDinger
Open-source AI trading OS for Python strategies, backtesting, and execution.
Build Python strategies, backtest, and trade across crypto, stocks, and forex on your own stack.
QuantDinger is a self-hosted platform combining AI research, Python strategy development, and backtesting. It suits developers and traders wanting full control over their market data, strategy code, and broker credentials.
What it is
It executes server-side backtesting, multi-provider AI market analysis, and paper or live trading across crypto exchanges and traditional brokers. The v5 architecture separates the HTTP API, trading workers, scheduling, and Celery tasks into distinct container processes.
Who it's for
It suits independent Python developers and small trading teams who need a self-hosted automation stack with multi-tenant SaaS capabilities. It does not suit non-technical users looking for a plug-and-play signal service or automated black-box trading.
Key features
Strategy API V2
Write and deploy custom Python indicators and trading strategies directly.
Process Separation
Run independent containers for trading workers, schedulers, and Celery job execution.
Multi-Tenant SaaS
Launch your own trading platform with built-in user management, billing, and payments.
Agent & MCP Access
Interact with your trading stack through web interfaces, mobile H5, human APIs, and MCP.
Up and running in 3 steps
Our honest verdict
What's great
- Local-first design keeps market data, code, and broker credentials under your control
- Supports paper and live execution across both crypto exchanges and traditional brokers
- PostgreSQL-backed state with durable workers, audit logs, and optional Prometheus monitoring
- Includes built-in user management, billing, and payments for launching a multi-tenant SaaS
Where it falls short
- Requires managing a complex multi-container Docker Compose stack with PostgreSQL, Redis, and worker processes
- Production deployments need manual secret generation and careful environment configuration across multiple .env files
- High-risk configuration requires strict adherence to security guidelines regarding file permissions and read-only mounts
Pricing & license
QuantDinger is free under its Apache-2.0 licence, with the only cost being the server infrastructure required to run the self-hosted stack yourself.
Open remote AI-training roles
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Does QuantDinger provide investment advice?
No, the project does not provide investment advice and you must review your own risk and compliance requirements.
How are long-running strategies handled?
Long-lived strategy runtimes and pending orders are owned by the dedicated trading worker process, separate from finite Celery jobs.
