Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
The RL Bridge for LLM-based Agent Applications. Made Simple & Flexible.
| Date | Stars |
|---|---|
| 2026-07-24 | 5596 |
| 2026-07-25 | 5599 |
| 2026-07-28 | 5599 |
| 2026-07-30 | 5599 |
| 2026-07-31 | 5620 |
| 2026-08-10 | 5651 |
| 2026-08-18 | 5672 |
| 2026-08-19 | 5679 |
| 2026-08-20 | 5680 |
| 2026-08-21 | 5684 |
| 2026-08-22 | 5685 |
| 2026-08-23 | 5686 |
| 2026-08-24 | 5687 |
| 2026-08-25 | 5692 |
| 2026-08-26 | 5695 |
| 2026-08-27 | 5699 |
| 2026-08-28 | 5699 |
| 2026-08-29 | 5702 |
| 2026-08-30 | 5704 |
| 2026-08-31 | 5705 |
| 2026-09-01 | 5706 |
| 2026-09-02 | 5711 |
| 2026-09-03 | 5716 |
| 2026-09-04 | 5721 |
| 2026-09-05 | 5728 |
| 2026-09-06 | 5728 |
| 2026-09-07 | 5733 |
| 2026-09-08 | 5737 |
| 2026-09-09 | 5740 |
| 2026-09-10 | 5749 |
| 2026-09-12 | 5750 |
| 2026-09-13 | 5751 |
| 2026-09-14 | 5758 |
| 2026-09-15 | 5762 |
| 2026-09-16 | 5765 |
| 2026-09-17 | 5768 |
| 2026-09-18 | 5774 |
| 2026-09-19 | 5776 |
| 2026-09-20 | 5780 |
Today
+4 stars today
This week
+29 stars this week
This month
+96 stars this month
Momentum
80.0
growth rate 0.50%/day
<h1 align="center"> <em>AReaL</em>: A Large-Scale Asynchronous Reinforcement Learning System </h1> <p align="center"> | <a href="https://arxiv.org/pdf/2505.24298"><b>Paper</b></a> | <a href="https://areal-project.github.io/AReaL/"><b>Documentation</b></a> | <a href="https://areal-project.github.io/AReaL/zh/"><b>中文文档</b></a> | <a href="https://deepwiki.com/areal-project/AReaL"><b>Ask DeepWiki</b></a> | <a href="https://huggingface.co/collections/inclusionAI/"><b>🤗 Models & Data</b></a> | <a href="./assets/figures/wechat_qrcode.png" target="_blank"><img src="./assets/figures/wechat_icon.png" width="20" style="vertical-align: middle;"> <b>WeChat (微信) Group</b></a> | <a href="https://www.bestpractices.dev/projects/12770"><img src="https://www.bestpractices.dev/projects/12770/badge"></a> </p> <img align="right" alt="ReaL" src="/assets/figures/logo.png" width="20%"> **AReaL** is a reinforcement learning (RL) infrastructure designed to bridge foundation model training with modern agent-based applications. It was originally developed by researchers and engineers from Tsinghua IIIS and the AReaL Team at Ant Group. Built on a **fully asynchronous RL training paradigm**, AReaL is optimized for efficiency and scalability, making it particularly well-suited for training large-scale **reasoning and agentic models**. AReaL’s mission is to make building AI agents **accessible, efficient, and cost-effective** for a broad community of developers and researchers. > Like milk tea - customizable, scalable, and enjoyable - we hope AReaL brings both > flexibility and delight to your AI development experience. Cheers! **AReaL Highlights** - ⚡ **Flexibility**: Seamless customization for [agentic RL](https://areal-project.github.io/AReaL/en/tutorial/agentic_rl.html) and [online RL training](https://areal-ai.io/en/tutorial/online_proxy.html) for **black-box agent applications** by simply replacing the `base_url`. - 📈 **Scalability**: **Stable** fully asynchronous RL training with **industry-leading speed**. - ✨ **Cutting-Edge Performance**: State-of-the-art [math](/blog/AReaL_v0_2.md), [coding](/blog/AReaL_v0_3.md), [search](https://github.com/inclusionAI/ASearcher), and [customer service](https://arxiv.org/abs/2601.22607) agents. ## 📰 News **\[2026/07/01\]** 🎉 We release **AReaL 2.0**, a major architectural milestone. This release refactors AReaL into a **microservice architecture** with independent [training](./areal/v2/training_service/), [inference](./areal/v2/inference_service/), [agent](./areal/v2/agent_service/), and [weight-update](./areal/v2/weight_update/) services, and ships the [Hermes online RL loop](./examples/hermes) and [end-to-end SWE RL training examples](./examples/swe). Check out our [technical report](https://arxiv.org/abs/2607.01120). **\[2026/06/17\]** 🔬 Introducing **KPop** — bidirectional binary KL divergence token masking. Configured via `rejection_sampling.metric=binary_kl`. Also adding an [IcePop config](examples/math/gsm8k_icepop.yaml) (importance-ratio-based token masking). Check out [gsm8k_kpop.yaml](examples/math/gsm8k_kpop.yaml) and [gsm8k_icepop.yaml](examples/math/gsm8k_icepop.yaml) to get started! <details> <summary><b>📋 Previous Releases</b></summary> **\[2026/04/23\]** 🚀 We’re excited to release our integration with [Scaffoldings](https://github.com/NVIDIA/TensorRT-LLM/tree/main/tensorrt_llm/scaffolding) for agentic RL training - now live in our [examples](https://github.com/areal-project/AReaL/tree/main/examples/scaffolding)! Huge shoutout to @narutolhy and @WeiHaocheng for making this happen 🙌. The modular design of the Scaffoldings enables it to achieve a thorough decoupling of agent execution, reward calculation, and trajectory acquisition. This enables developers to reuse existing modules when implementing an agentic RL method, allowing them to focus on their own innovative modules. **\[2026/04/18\]** We are thrilled to announce that **AReaL's first Community Biweekly Meeting
Excerpt of 27,720 characters
Read on GitHub
400
Xujie Shen
45
29
26
24
ZIYI ZENG · CUHKSZ · China
21
19
Rongzhi Gu · Huawei Technologies Canada · Canada
14
zhangjw
14
13
11
10
9
9
8
7
6
6
6
Maxwill Lin · vmax.ai · United States
5
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:a9b1b2e3fc5619b7, topic:reinforcement-learning, topic:rl, readme:reinforcement learning
matched fp:a9b1b2e3fc5619b7, topic:llm, readme:technical report
matched fp:a9b1b2e3fc5619b7, topic:llm-agent, readme:ai agents, readme:agentic