Freqtrade
github.com
Free, open-source Python crypto trading bot you host yourself, with backtesting, optimisation, machine-learning strategies, dry runs and Telegram control.
What it is
Freqtrade is a self-hosted bot for trading crypto on exchanges. You write a strategy as a Python class, download historical candles, backtest it, tune its parameters with the built-in hyperopt optimiser, then run it in dry-run mode with simulated money before letting it trade a live account. It supports spot markets on Binance, BingX, Bitget, Bybit, Gate, HTX, Hyperliquid, Kraken and OKX, plus others through the CCXT library, and leveraged futures on a smaller list of exchanges.
FreqAI, an optional module, lets strategies train and retrain machine-learning models on market data as the bot runs. Running bots are controlled through Telegram commands or FreqUI, a browser dashboard that shows open trades, profit and backtest results.
The software itself is free with no paid tier. The costs are the ones you bring: a server or always-on computer (the documentation suggests at least 2 GB of RAM and 2 vCPUs), your exchange's trading fees and your time. It is aimed at people comfortable with Python, Docker and the command line; there is no hosted version or support desk, only documentation and a community Discord. The project states it is for educational use and that you trade at your own risk.
Key features
- Backtesting
- Replay strategies over downloaded historical data with reports on profit, drawdown and trade statistics.
- Hyperopt
- Machine-learning-based search for strategy parameters such as entry thresholds, stop losses and ROI targets.
- FreqAI
- Adaptive modelling that trains and retrains predictive models on live market data.
- Dry-run mode
- Runs the strategy against live prices with simulated funds before real money is used.
- Exchange support
- Spot trading on nine named exchanges and others through CCXT; futures on Binance, Bitget, Bybit, Gate, Hyperliquid, Kraken and OKX.
- Telegram and FreqUI
- Control and monitor the bot from Telegram or a built-in web interface.
- Pair lists
- Dynamic whitelists and blacklists choose which markets the bot trades.
Pricing
| Plan | Price | What you get |
|---|---|---|
| Open source | $0 | Full software; you supply hosting and pay exchange fees |
No paid version. Running it needs a machine left on, such as a small cloud server.
Free option: Everything is free and open source: backtesting, optimisation, FreqAI, dry-run and live trading.
Checked September 25, 2026 on github.com. Prices change; confirm on the provider's site before you subscribe.
Strengths
- Free, with the full source code available to inspect and change
- Complete workflow from backtest to optimisation to dry run to live trading
- Supports many exchanges through CCXT, including futures on several
- Large, active open-source community on GitHub and Discord
Limitations
- Crypto only; it does not trade stocks, options or other assets
- Requires Python or Docker knowledge and a machine that stays online
- No official support desk; help comes from documentation and the community
- Several supported exchanges do not accept US residents, so US users must check their exchange first
- The project describes itself as educational software used at your own risk
Good fit for
- Developers who want to build and test their own crypto strategies
- Traders who prefer self-hosting to handing API keys to a cloud bot service
- Quants experimenting with machine-learning strategies
Look elsewhere if
- Stock and options traders
- Anyone who wants a no-code, hosted bot
Compare
Alternatives to Freqtrade
Freqtrade questions
Is Freqtrade free?
Yes. It is open-source software with no paid tier; you only pay for hosting and exchange fees.
Which exchanges does Freqtrade support?
Spot trading on Binance, BingX, Bitget, Bybit, Gate, HTX, Hyperliquid, Kraken, OKX and others via CCXT, and futures on Binance, Bitget, Bybit, Gate, Hyperliquid, Kraken and OKX.
Can I test Freqtrade without risking money?
Yes. Backtesting uses historical data, and dry-run mode trades live prices with simulated funds.
Do I need to know Python to use Freqtrade?
Strategies are written in Python, and setup uses Docker or a Python environment, so some programming knowledge is expected.
Written from Freqtrade's own website and documentation. All Trading Tools lists and compares tools; it does not recommend trades or give financial advice. Sources: github.com/freqtrade/freqtrade, freqtrade.io/en/stable.