---
title: "Xiaohongshu Open Sources Large Model Reinforcement Learning Training Engine Relax"
type: "News"
locale: "en"
url: "https://longbridge.com/en/news/282779116.md"
description: "On April 15, the Xiaohongshu AI Platform Team officially open-sourced Relax, a large model reinforcement learning training engine designed for full-modal and Agentic scenarios. Relax supports unified processing and flexible integration of inputs such as images, audio, and video, combining modal-aware parallelism with end-to-end asynchronous pipelines to enhance the efficiency and scalability of multi-modal training"
datetime: "2026-04-15T03:49:39.000Z"
locales:
  - [zh-CN](https://longbridge.com/zh-CN/news/282779116.md)
  - [en](https://longbridge.com/en/news/282779116.md)
  - [zh-HK](https://longbridge.com/zh-HK/news/282779116.md)
---

# Xiaohongshu Open Sources Large Model Reinforcement Learning Training Engine Relax

On April 15, the Xiaohongshu AI Platform Team officially open-sourced Relax—a large model reinforcement learning training engine designed for full-modal and Agentic scenarios. Introduced as supporting unified processing and flexible integration of inputs such as images, audio, and video, it combines modal-aware parallelism with end-to-end asynchronous pipelines to improve the efficiency and scalability of multi-modal training.

### Related Stocks

- [XHS.NA](https://longbridge.com/en/quote/XHS.NA.md)

## Related News & Research

- [Advancing Open Source Patent Protection: Preservation of OIN 2.0 Source Code](https://longbridge.com/en/news/286806696.md)
- [Open source tool maker Grafana Labs says hackers stole its code, refuses to pay ransom](https://longbridge.com/en/news/286782006.md)
- [14:39 ETClarityCheck Announces 25M+ Total Searches Since Launch](https://longbridge.com/en/news/286454452.md)
- [TG-17 CEO Featured Live on Big Biz Show](https://longbridge.com/en/news/286430085.md)
- [Raiffeisen Raises Addiko Bid 15% As Rival 29 Offer Emerges](https://longbridge.com/en/news/286485742.md)