# Ant Ling — AReaL: Ant Reasoning Reinforcement Learning for LLMs

- Company: Ant Ling (ant-ling.com)
- Announced: 2025-04-01
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://developer.ant-ling.com/en/blogs/areal
- Record: https://forck.live/items/22344-areal-ant-reasoning-reinforcement-learning-for-llms
- Subject: Ling / Ring / Ming
- Open weights: yes
- Models affected: Qwen3, AReaL-boba²-8B, AReaL-boba²-14B, AReaL-boba²-8B-Open, AReaL-boba²-14B-Open

Ant Research released AReaL (Ant Reasoning RL), an open-source fully asynchronous reinforcement learning training system for large reasoning models. The v0.3 release (boba²) achieves a 2.77x speedup over synchronous systems and includes SOTA coding models: a 14B model scoring 69.1 on LCB-v5. The system supports multi-turn agentic RL training and scales from a single node to 1K GPUs.

## Evidence

Verbatim from https://developer.ant-ling.com/en/blogs/areal:

> AReaL (Ant Reasoning RL) is an open-source fully asynchronous reinforcement learning training system for large reasoning models developed at the RL Lab, Ant Research . Built upon the open-source project RealHF , we are fully committed to open-source by providing training details, data, and infrastructure required to reproduce results along with the model itself.

---

Record: https://forck.live/items/22344-areal-ant-reasoning-reinforcement-learning-for-llms
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md