Trading

ccxt43.7k★

CCXT gives you one API for over 100 crypto exchanges instead of 100 different APIs. Every exchange has its own authentication, its own rate limits, and its own idea of what an order book looks like. CCXT normalizes all of it behind a single interface so the same code runs against different venues and prediction markets. MIT licensed, shipping for Node 18+, Python 3, PHP 8.1+, .NET, Go 1.20+, Java 21+, and browsers. There is no infrastructure to run. It is a library you install and call, which means the operational work is all yours: key management, rate limiting, retry logic, and handling the exchange-specific failures that normalization cannot hide. That is the right trade for most people, but it is not zero effort just because there is no server. The free library covers REST, which is enough for backtesting, portfolio tracking, and any strategy that polls. Real-time WebSocket streaming, meaning live order books, trades, and balance updates, is CCXT Pro, a paid product built on top of the free library. They do not publish the price, so you have to ask. The catch is that a unified API is a leaky abstraction. Exchanges differ in ways normalization cannot paper over: fee structures, order types, margin rules, withdrawal behavior. CCXT gets you to a working prototype fast, then you spend the real time on per-exchange edge cases. Worth saying plainly: this is a trading library, not a trading strategy, and it will execute a bad idea as cheerfully as a good one.

nautilus_trader26.7k★

NautilusTrader is an open source engine for building and running algorithmic trading systems. Strategy logic lives in Python, and a compiled Rust core underneath handles event processing and order execution with nanosecond-resolution timing. The same strategy code runs in backtest and in live trading, so what you tested is what you ship. LGPL-3.0, nothing in the repository is gated. That research-to-live parity is the reason to pick it over a backtest-only framework, and it is the failure mode that kills most homegrown systems. Setup is not a weekend. You supply market data, wire up broker or exchange adapters, and run your own infrastructure for anything live. The project is also mid-migration to a Rust-native v2 runtime with PyO3 bindings, and v1 now gets critical security backports only, so start new work on v2. Solo quants and small teams who can operate their own boxes are the audience. If all you need is exchange connectivity, ccxt (ccxt/ccxt) is the much smaller building block. Nautilus is what you pick when the accuracy of the fill model is the thing you are betting on. The catch: the free ride may be ending at the edges. Nautech has announced a Pro subscription of Docker images adding a live dashboard, prebuilt execution algorithms, and sub-microsecond IPC, plus a managed Cloud Platform and an Institutional tier. All three say coming soon and none publishes a price. The core engine is still complete and still LGPL, but plan on the good operational tooling living behind a subscription.

turboquant_plus7.0k★

TurboQuant Plus is a quantitative trading platform for building and backtesting trading strategies. Python-based with support for multiple data sources and strategy templates. If you're developing algorithmic trading systems, this provides the framework for strategy development, historical backtesting, and performance analysis. Fully free, no paid tier. The catch: documentation and community are primarily in Chinese. Quantitative trading tools require significant domain expertise in both finance and programming to use effectively. No English documentation means you need Chinese language ability or patience with translation tools. And backtesting performance doesn't guarantee live trading results. The trading platform space has established alternatives like Zipline and Backtrader with English-language communities.