redtorch.io

What It Is

RedTorch is an open-source quantitative trading application development framework based on Kotlin/Java. Its repository is hosted on GitHub and it is released under the MIT license. According to the project description, version 2.0.0 has been refactored and streamlined: it no longer aims to be an all-in-one quant trading system, but is instead positioned as a core trade management system focused on unified management and monitoring of data such as accounts, positions, orders, fills, contracts, and market data subscriptions.

Core Capabilities and Tech Stack

The project has been rewritten in Kotlin while remaining compatible with the Java ecosystem. It requires JDK 17+, JavaFX 17+, and Gradle 7+. Its modules include rt-master as the master node, rt-slave as the slave node, rt-desktop as the GUI terminal, rt-gateway-api as the gateway interface, and rt-gateway-ctp as the CTP gateway implementation. It depends on Zookeeper, and data stored in Zookeeper is additionally desensitized with 3DES encryption. Protobuf has been removed for data transmission, with RFC 6902 now used for differential synchronization. Developers can extend the gateway interface to connect different underlying APIs; it is not limited to futures, although scenarios such as securities trading require custom adapters.

Pricing and Deployment

RedTorch is free and open source, and no paid edition was found. The project supports self-hosting: you need to run Zookeeper yourself, then start each module via its main method, or build a jar with Gradle bootJar and run it with java -jar. No additional container is required. The application.properties configuration file can override the built-in configuration from the runtime directory, which makes production deployment easier.

Pros and Cons

Its advantages are that it is open source, has a clear architecture, and is straightforward to deploy. Version 2.0.0 also removes non-core features such as web administration, web trading pages, historical data, and candlestick calculation, reducing both the attack surface and overall complexity. The drawbacks are also clear: the current version has not been fully tested, and the official guidance explicitly says it must be reviewed and tested before use; the project provides no documentation beyond an FAQ; and the author does not offer commercial support. The Python client is also no longer supported, making it less friendly for users outside the Java/Kotlin ecosystem.

Who It’s For and Access from China

RedTorch is suitable for quant developers with Java/Kotlin skills who want to build their own trade management system or study CTP gateway integration. It is not suitable for simulated-account/sub-account position-splitting use cases, high-frequency trading that requires Tick-to-Order latency below 3 milliseconds, or users looking for an all-in-one quant trading platform. Regarding access from China, the source text does not provide availability or payment information for redtorch.io or GitHub, so this is unknown. Alternatives to compare include vn.py, Backtrader, Zipline, and QuantConnect.