selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

Freqtrade logo

Self-host Freqtrade

2026.10

Open-source crypto trading bot with backtesting and ML

ReleaseOpen SourceGPL-3.0
Docker ComposeCompose file & guide
docker compose up -d
Updated2 weeks ago
LanguagePython
Setup difficultyModerate
MobileResponsive
ThemeDark mode
Stars
56 945
CPU
2vCPU
RAM
2GB RAM
Storage
1GB disk space
Port
8080 (REST API / freqUI, bound to localhost by default in the shipped docker-compose (host mapping 127.0.0.1:8080:8080))
Docker
freqtradeorg/freqtrade
Overview screenshot of the freqtrade bot shown in the READMEfreqUI login screenfreqUI trade view with open tradesfreqUI settings viewfreqUI backtesting view comparing results

What is Freqtrade?

Freqtrade is a free and open-source Python bot for automated cryptocurrency trading, aimed at users comfortable writing and testing their own strategies in Python. It connects to major spot and futures exchanges (Binance, Kraken, OKX, Bybit, Bitget, Gate, HTX, Hyperliquid and others via ccxt), keeps its trade history in a local SQLite file, and can simulate trading in dry-run mode before any real money is involved. Strategies are Python classes that can be validated with the built-in backtesting engine, tuned through a machine-learning hyperopt module, or made self-adapting with FreqAI, which continuously retrains models on live market data. A built-in REST API and the bundled freqUI web interface let you monitor positions, balances and performance from the browser, and a Telegram bot exposes commands to start, stop and inspect the trader remotely. Webhooks and a producer/consumer data stream let you push bot messages elsewhere. Installation is documented with a Docker Compose quickstart, native setup scripts for Linux/macOS (./setup.sh) and Windows (./setup.ps1), a Conda path, and Raspberry Pi notes; the project recommends an NTP-synced clock and at least 2 GB of RAM, and stresses that results are not guaranteed and software is used at your own risk.

Freqtrade alternatives

Should you self-host Freqtrade?

Freqtrade scores 87/100 on selfhost.directory (how scoring works) and ranks #1 of 44 in Cryptocurrency. It has 56 945 GitHub stars, +1 272 in the last 30 days.

It shipped 11 stable releases in the past year. The latest, 2026.10, came out on 30 October 2026. New versions land roughly every 4 weeks.

Setup is moderate: expect some configuration beyond a single command. An official Docker image is available. It is GPL-3.0-licensed and written mostly in Python.

Its closest alternative on selfhost.directory is Bitcoin Core (86/100, 90 691 stars). Freqtrade scores higher overall. See Freqtrade vs Bitcoin Core side by side.

Features

  • Adaptive ML strategies (FreqAI)Dedicated docker image tags exist for FreqAI, PyTorch and reinforcement learning.
  • Backtesting
  • Built-in web UI
  • Dry-run / paper trading
  • Dynamic pair whitelist/blacklist
  • Fiat profit display
  • Performance reporting
  • Plotting / chartingRequires the *_plot docker image variant.
  • Producer/consumer data streaming
  • REST API
  • Short sellingFutures trading supported only for selected exchanges.
  • SQLite persistence
  • Strategy optimization by machine learning
  • Telegram control
  • Webhook Notifications

Freqtrade README

freqtrade

Freqtrade CI DOI codecov Documentation Discord Server

Freqtrade is a free and open source crypto trading bot written in Python. It is designed to support all major exchanges and be controlled via Telegram or webUI. It contains backtesting, plotting and money management tools as well as strategy optimization by machine learning.

freqtrade

How to install Freqtrade

Install Freqtrade with Docker Compose

Docker
mkdir ft_userdata
cd ft_userdata/
# Download the docker-compose file from the repository
curl https://raw.githubusercontent.com/freqtrade/freqtrade/stable/docker-compose.yml -o docker-compose.yml
# Pull the freqtrade image
docker compose pull
# Create user directory structure
docker compose run --rm freqtrade create-userdir --userdir user_data
# Create configuration - Requires answering interactive questions
docker compose run --rm freqtrade new-config --config user_data/config.json

Then launch with docker compose up -d.

Official image on Docker Hub; recommended for Raspberry Pi and ARM64 systems.

How to configure Freqtrade

  • max_open_tradesRequired. Number of open trades your bot is allowed to have. Only one open trade per pair is possible, so the length of your pairlist is another limitation that can apply. If -1 then it is ignored (i.e. potentially unlimited open trades, limited by the pairlist). More information below. Strategy Override. Datatype: Positive integer or -1.
  • stake_currencyRequired. Crypto-currency used for trading. Datatype: String
  • stake_amountRequired. Amount of crypto-currency your bot will use for each trade. Set it to "unlimited" to allow the bot to use all available balance. More information below. Datatype: Positive float or "unlimited".
  • tradable_balance_ratioRatio of the total account balance the bot is allowed to trade. More information below. Defaults to 0.99 99%). Datatype: Positive float between 0.1 and 1.0.
  • available_capitalAvailable starting capital for the bot. Useful when running multiple bots on the same exchange account. More information below. Datatype: Positive float.

Freqtrade releases & version history

@ Every self-hosted release, weekly

One Monday email with the week's stable releases across the directory.

FAQ

I want to use incomplete candles

Freqtrade deliberately only provides completed candles to strategies; for current market data it offers orderbook and ticker methods via the dataprovider, which cannot be backtested.

I deposited more funds to the exchange, but my bot doesn't recognize this

Freqtrade refreshes balances before placing orders; RPC/API balance calls update at most once per hour, and /reload_config forces an immediate refresh.

Why does my bot not sell everything it bought?

Exchange fees taken from the received currency leave "coin dust" below the minimum tradable lot size; this also happens in manual trading, and paying fees in a dedicated fee currency (e.g. BNB) can fix it where supported.

I'd like to make changes to the config. Can I do that without having to kill the bot?

Yes — edit the config and use the /reload_config command, which stops the bot, reloads configuration and strategy, and restarts with them.

I have made 12 trades already, why is my total profit negative?

A dozen trades is too small a sample to judge; the results only become meaningful over thousands of trades, and losses on specific pairs are normal.

I have waited 5 minutes, why hasn't the bot made any trades yet?

Depending on the entry strategy, pairlist and market conditions it can take hours or days to find an entry; check the logs to see whether the bot simply gets no signals.

The bot starts, but in STOPPED mode

Set the initial_state config option to "running" in your config.json.

The bot does not start

A "command not found" error usually means the Python virtual environment is not active or the installation did not complete; activate the venv and check the installation docs.

Does freqtrade support sandbox accounts?

No — exchange sandbox markets are considered unrealistic; use the built-in dry-run mode to simulate trading without real funds.

Can freqtrade open multiple positions on the same pair in parallel?

No, it opens only one position per pair at a time, though the adjust_trade_position() callback can adjust an open position.

Can my bot trade options or futures?

Futures trading is supported for selected exchanges; the docs' start page lists the supported ones.

Can my bot open short positions?

Yes, on futures markets, provided the strategy supports it and "trading_mode": "futures" is set. On spot markets, inverted leveraged spot tokens can sometimes be traded.

More in Cryptocurrency

Rate this project

Click to rate — no account needed.

0 comments

Posting as guest
Loading comments…