Quantify is a free, open-source trading terminal written in Python. It is aimed at traders who want to test short-term and multi-strategy ideas on crypto, A-shares and US stocks before risking money, and at developers who prefer to read and modify the code behind their tools. Its interface imitates a familiar exchange screen, with a dark order-book-style layout, live candles and a signal panel.
The program runs as a local web application: you install the Python dependencies, start the server, and open the interface in your browser at localhost. Because it is licensed under MIT, you can use, copy and change it freely, including for commercial work. Its most distinctive feature is honesty about results: it is built to show whether a strategy actually has an edge after fees and sample-size limits, rather than to flatter it.
What Is Quantify and Who Is It For?
Quantify is a research and analysis environment rather than a finished product that promises returns. It brings together charting, live market data, indicator calculations, strategy signals and backtesting in one local application. You run it on your own computer, so the data and your strategy code stay with you.
The most natural users are active traders who already read charts and want to check their ideas numerically before acting on them. Quantitative hobbyists and students of market structure will also find it useful, because every strategy is a readable Python component that can be inspected and adjusted.
It is less suited to people looking for a set-and-forget money machine. The software can show you what a strategy did in the past and how many trades it took to get there, but it cannot guarantee that the same pattern will repeat.
How Backtesting Helps Separate Ideas From Evidence
A backtest replays a strategy over historical prices and records what the trades would have produced. This is useful because it turns a vague belief such as "RSI bounces work on this coin" into a set of numbers: how many trades, what average gain or loss per trade, and how much fees reduced the result.
Quantify emphasizes out-of-sample comparison. The idea is to tune a strategy on one stretch of data and then check it on a separate period it has not seen. A strategy that looks excellent in-sample but collapses out-of-sample has probably been fitted to noise.
Sample size matters as much as the average result. Ten trades can produce any number, which is why the program greys out performance figures when the sample is too small. Treat that warning seriously: a short winning streak is not evidence of an edge.
Understanding Expectancy, Reward-to-Risk and Fees
Expectancy is the average amount you can expect to make or lose per trade, calculated from the win rate and the average win and loss. A strategy can win most of its trades and still have negative expectancy if its losses are larger than its gains.
Reward-to-risk compares the size of the typical target to the typical stop. A system that risks two units to make one can still be profitable, but only if it wins often enough. Short-term strategies are especially sensitive to fees, because frequent trading multiplies the cost of every entry and exit.
This is why Quantify's short-term health check is useful. It shows whether the gross edge survives after commissions, which is the question that many enthusiastic backtests never answer.
Practical Use Cases
One common workflow is to take a familiar idea, such as a moving-average crossover or a Bollinger bounce, and run it on several timeframes and symbols. The results show where the idea works and where it only looked promising by chance.
Another is building a simple decision checklist. When several strategies agree on a direction, the consensus view can serve as a filter for manual trades. When they disagree, the desk shows that the signal is mixed, which is itself useful information.
Developers can also use Quantify as a base for their own work. Because each strategy is a self-contained Python component, adding a new rule or changing a parameter does not require rewriting the interface. Researchers comparing data sources can switch between crypto, A-share and US-stock feeds with a single configuration change.
Key features
Exchange-style candlestick terminal
Live crypto prices over WebSocket
Wide set of overlay and oscillator indicators
Multi-strategy decision desk
Nine pluggable strategies
Short-term health check and robustness tests
One interface for three markets
Transparent warnings
Supported coins
Documentation mentions Bitcoin (BTC), Ethereum (ETH), Dogecoin (DOGE), XRP, BNB Chain tokens and USDT as examples of supported assets. Crypto data is handled through ccxt and the gate WebSocket feed, while A-shares and US stocks are supported through the Sina and Yahoo data sources respectively. Check the repository for the current list of exchanges and symbols.
How to install Quantify on Windows
- 1
Get the source code
Download the project from the official GitHub repository at github.com/Zhanghanser/quantify, either by cloning it with Git or by downloading the archive from the repository page. Use only this official source, and check the license file before reusing the code.
- 2
Install Python and dependencies
Install a recent version of Python for Windows from python.org, making sure to tick the option to add Python to PATH. Open a terminal in the project folder and run pip install -r requirements.txt to install the libraries the application needs.
- 3
Start the local server
In the same terminal, run python runweb.py. The server starts on your own machine and prints its address in the console.
-
Open the interface
Open your browser and go to http://localhost:8000. The terminal should load with its charts and panels. If the page does not open, check the console for error messages about missing packages.
System requirements
- Windows with Python installed, as the program is run from the Python environment
- Internet connection for live crypto and market data; without it, the program falls back to simulated data
How to use Quantify
Quantify runs in your browser, so the workflow starts with choosing a market and a symbol and ends with reading the decision panel. The steps below describe a typical session.
- Start the server with python runweb.py and open http://localhost:8000 in your browser.
- Select a market and symbol. For crypto, the live feed begins immediately; for stocks, make sure your data source is set in the configuration.
- Choose a timeframe and add the indicators you want, such as MA, Bollinger Bands on the main chart and RSI or MACD in a sub-panel.
- Open the strategy list and enable the strategies you want to compare. Start with one or two so the results are easy to interpret.
- Run a backtest and read the results, paying attention to the number of trades and whether the out-of-sample figures agree with the in-sample ones.
- Check the short-term health metrics, especially expectancy after fees, before drawing any conclusion.
- Review the decision panel's suggested order, including stop-loss and position size, as a planning aid rather than an instruction.
Pros & cons
- Free and open source under the MIT license
- Readable Python code that can be inspected and extended
- Nine built-in strategies, including five short-term approaches
- Real-time crypto data through a direct WebSocket connection
- Built-in honesty checks such as fee drag and small-sample warnings
- One interface for crypto, A-share and US-stock data
- Requires comfort with installing Python packages and running a command-line server
- Documentation in the repository is largely in Chinese, which may slow English-speaking users
- Results depend heavily on the quality and length of the historical data used
- Not a turnkey service; you must configure data sources and manage the environment yourself
- Live trading execution is not the focus, as the decision desk is signal-only
Is Quantify safe?
Quantify is a set of Python scripts, so treat the code the way you would any downloaded program. Get it only from the official GitHub repository, check the commit history and license, and review the requirements file before installing packages. Verify the integrity of any release archive with its published SHA-256 checksum where one is provided. Install dependencies inside a virtual environment so that a problem in one project cannot affect the rest of your system. Never enter exchange API keys or wallet seed phrases into third-party pages that claim to be part of the project; keep credentials in local configuration files you control. Antivirus tools sometimes flag scripts that access the network or run local servers, so inspect the code rather than disabling protection. Note that the project status and last code update should be checked before relying on it for anything important.
- Official source only. Download from github.com/Zhanghanser/quantify. Ignore ads and lookalike domains.
- Check the SHA-256. Compare the installer hash with the value the developer publishes before running it.
- Never share your seed phrase. No developer, support agent or website — including CryptApps — will ever ask for it.
Alternatives
Several of the related programs overlap with Quantify, but each is built for a different purpose. Hummingbot focuses on market making and arbitrage with exchange connectors, while OctoBot is a trading bot with a web interface designed to run strategies continuously. Gunbot Quant and Tradingview MCP also lean toward analysis and screening. StockSharp is a broader algorithmic trading platform written in C#, suited to developers building systems across many markets.
Quantify's distinguishing trait is its combination of a chart-first terminal with explicit honesty tools. It is less about running a bot around the clock and more about checking whether an idea deserves money in the first place. Users who want execution automation may prefer OctoBot or Hummingbot, while those who want to validate short-term logic before automating may find Quantify a better starting point. Gekko and Catalyst are archived projects and are best considered for historical reference.
Specifications
- Price
- Free
- Type
- Open source
- License
- MIT
- Latest version
- —
- Last update
- 30 Jun 2026
- Windows support
- Yes (Windows build or installer)
- Source code
- github.com/Zhanghanser/quantify
Frequently asked questions
Yes. Quantify is released under the MIT license and is free to download, use, modify and redistribute. The license allows commercial use as well, provided the copyright notice is kept. Check the repository's license file for the exact terms before reusing the code.
The decision desk is described as signal-only, meaning it suggests orders with stop-loss, take-profit and position size but is not designed as an automated execution engine. It helps you evaluate and plan trades. Any step toward live execution requires your own additional development and careful testing.
Quantify supports crypto markets through ccxt and a direct gate WebSocket feed, A-shares through Sina, and US stocks through Yahoo. Switching between them is a configuration change. Availability of specific exchanges and symbols can change, so check the repository for the current list.
Install Python, download the project from the official GitHub repository, and run pip install -r requirements.txt in the project folder. Then start the server with python runweb.py and open http://localhost:8000 in your browser. The interface runs locally on your own computer.
The software is open source, so its code can be reviewed, which is an advantage. Download it only from the official repository, keep your API keys out of third-party sites, and use a virtual environment for dependencies. Review the code before trusting it with real accounts, since you are responsible for what it runs.
No tool can guarantee profits, and Quantify does not claim to. Its purpose is to test strategies against historical data and show their expectancy, fee impact and sample size. A backtest that looks good may not repeat in live markets, so treat results as a starting point for further testing.
The warnings are intentional. If the network is down and simulated data is used, a red label appears so you do not mistake fake prices for real ones. Metrics based on too few trades are greyed out, and binary-options mode displays the negative long-term expectancy of those products.
The program includes nine pluggable strategies: dual moving averages, channel breakout, RSI mean reversion and a trend filter, along with five short-term approaches covering Bollinger bounce, fast RSI rebound, volume breakout, VWAP reversion and Bollinger-plus-RSI double confirmation. You can enable and compare them individually.
Verdict
Quantify is a strong choice for traders and developers who want an honest, readable, open-source environment for testing trading ideas across crypto and stock markets. Its free license, transparent code and built-in warnings about small samples and fee drag set it apart from tools that simply display attractive returns. It is best suited to people comfortable with Python and command-line setup. If your goal is fully automated execution, look at related projects such as OctoBot or Hummingbot. Always check the latest version and project status on the official repository before relying on it.