AKQuant

AKQuant is a Python and Rust framework that tests and runs trading strategies on stocks, futures and options, with an MCP server for AI assistants.

What it is good for

  • Backtesting a stock trading strategy on historical data
  • Running the same trading strategy in simulation or live
  • Testing trading rules that follow China A-share market rules
  • Asking an AI assistant to write and run a backtest

At a glance

Who it is for
Setup difficulty
Developer ยท Requires programming

Level 5 of 5. You need to write code to use it.

Cost
Free and open source
Type
Framework or SDK
Tasks
Works with
Claude, Codex, Cursor, AKShare, CTP futures interface, PyTorch

How to get started with AKQuant

  1. Install uv if you want to use it through an AI assistant.
  2. Tell your AI assistant to install the official AKShare and AKQuant MCP and run a sample strategy.
  3. Or install the library with pip install akquant (Python 3.10 or newer).
  4. Write a strategy by subclassing Strategy and run it with run_backtest.

From the project's own documentation. If a step does not match what you see, the official pages have the latest instructions.

Before you try it

  • Backtests by default charge no trading costs and allow same-day trades, so results look optimistic.
  • Writing strategies by hand requires Python code.
  • Manual MCP setup needs a project folder created beforehand and passed with `--root`.
  • Documentation and README are written mainly in Chinese.

Momentum

Rank this week
#41 in builder tools
Best rank
#41 week 46, 2026
Stars gained last week
+38
GitHub stars
2,610

GitHub numbers as of Nov 16, 2026. Updated weekly.

2,557 2,586 2,614 Nov 7, 2026 Nov 16, 2026
GitHub stars for AKQuant: 2,561 on Nov 7, 2026, 2,610 on Nov 16, 2026. Source: GitHub.

Ranking history

First ranked in week 46, 2026.

Questions

Is AKQuant free?
Yes. It is free and open source (MIT license).
Do I need to know how to code to use AKQuant?
Yes. You need to write code to use it.
What does AKQuant work with?
According to its own documentation, it works with Claude, Codex, Cursor, AKShare, CTP futures interface, and PyTorch.
How do I get started with AKQuant?
1. Install uv if you want to use it through an AI assistant. 2. Tell your AI assistant to install the official AKShare and AKQuant MCP and run a sample strategy. 3. Or install the library with pip install akquant (Python 3.10 or newer). 4. Write a strategy by subclassing Strategy and run it with run_backtest.
What should I know before trying AKQuant?
Backtests by default charge no trading costs and allow same-day trades, so results look optimistic. Writing strategies by hand requires Python code. Manual MCP setup needs a project folder created beforehand and passed with `--root`. Documentation and README are written mainly in Chinese.
What is AKQuant good for?
Backtesting a stock trading strategy on historical data, running the same trading strategy in simulation or live, testing trading rules that follow China A-share market rules, and asking an AI assistant to write and run a backtest.

Similar agents to AKQuant

  • AKTools

    Provides AKShare financial data to AI assistants, but does not run backtests or trading.

  • CNEquity

    Focuses on a local China A-share data lake for quant research rather than backtesting and live trading.

  • Vibe-Research

    A self-hosted stock research workbench with an AI agent, not a strategy backtesting framework.

  • a-stock-data

    A skill file for pulling A-share market data, without a backtest or trading engine.

  • HyperGrok Trading Desk

    Reviews Hyperliquid trades with read-only tools before orders, rather than testing strategies on history.

  • dbx

    dbx is a small desktop database client that lets you browse, query and edit data across 100+ databases, with an AI SQL helper and MCP server.

Last updated Nov 16, 2026. Tracked since Nov 7, 2026. Spotted an error? Tell us.