AReaL

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 and online RL training 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, coding, search, and customer service 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, inference, agent, and weight-update services, and ships the Hermes online RL loop and end-to-end SWE RL training examples. Check out our technical report.

[2026/06/17] 🔬 Introducing KPop — bidirectional binary KL divergence token masking. Configured via rejection_sampling.metric=binary_kl. Also adding an IcePop config (importance-ratio-based token masking). Check out gsm8k_kpop.yaml and gsm8k_icepop.yaml to get started!

[2026/04/23] 🚀 We’re excited to release our integration with Scaffoldings for agentic RL training - now live in our examples! 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 was successfully held! Thank you to everyone who joined us. Meeting materials are now available here. Our next meeting is scheduled for 2026/05/01 and will also be conducted in Chinese; English-language meetings will be scheduled in the future. We warmly welcome everyone to participate! See Community for more details.

[2026/03/02] We provide a complete example to train your own 🦞 OpenClaw agent by simply replacing the base_url and api_key with AReaL's RL service - no complicated dependencies, no code changes, works with any agentic runtime!

[2026/02/06] We are delighted to introduce AReaL-SEA, a self-evolving data synthesis engine. Combined with RL training on AReaL, the 235B MoE model surpasses GPT 5 and achieves comparable performance with Gemini 3.0 Pro on $\tau^2$-bench! Check out the paper, model, data, and code.

[2026/01/15] Congrats to our friends at CAMEL-AI for open-sourcing SETA, their terminal agent RL project trained with AReaL! Check out their training workflow and the announcement on X.

[2026/01/01] Happy New Year! Thanks to the outstanding contribution from @HwVanICI, we are excited to officially announce stable support for AReaL training on Huawei Ascend NPU devices! The code is actively maintained and continuously updated in the ascend branch. Check out our documentation to get started, and feel free to report any issues!

[2025/08/30] Introducing ASearcher, a state-of-the-art search agent built with AReaL's end-to-end asynchronous RL training. Check out the paper and the open-source repository!

[2025/07/31] (AReaL-lite) We introduce AReaL-lite, a lightweight version of AReaL designed specifically for AI researchers and rapid prototyping. AReaL-lite features an algorithm-first API design that prioritizes ease of use and algorithm development, while natively supporting fully asynchronous agentic RL. With 80% fewer lines of code, AReaL-lite maintains 90% of AReaL's performance and core functionality. Check out our documentation and the quickstart guide to begin your journey with AReaL-lite!

[2025/06/03] (v0.3, boba²) We release boba² (double-boba) for fully asynchronous RL training, which achieves 2.77× speedup while delivering comparable or superior training performance compared to synchronous systems. Furthermore, asynchronous RL significantly simplifies multi-turn agentic RL training setup! Check out our v0.3 overview blog and the research paper.

[2025/03/31] (v0.2, boba) Introducing our milestone release—boba! Please call it A-ReaL-boba! This release features significantly faster training with SGLang support and state-of-the-art 7B and 32B models for mathematical reasoning. Check out our v0.2 technical blog.

[2025/02/24] (v0.1) Our initial release includes reproducible results for 1.5B and 7B Large Reasoning Models (LRMs). Check out our v0.1 technical blog.

🚀 Getting Started

First, install the package:

git clone https://github.com/areal-project/AReaL
cd AReaL
pip install uv
# Install flash-attn pre-built wheel first to avoid compiling from source
# (pick the wheel matching your Python version; see https://github.com/mjun0812/flash-attention-prebuild-wheels/releases)
uv pip install "https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.16/flash_attn-2.8.3+cu128torch2.9-cp312-cp312-linux_x86_64.whl"
uv sync --extra cuda  # installs training packages + SGLang (default inference backend)
# For vLLM instead: cp pyproject.vllm.toml pyproject.toml && cp uv.vllm.lock uv.lock && uv sync --extra cuda

Our training scripts automatically download the required dataset (openai/gsm8k) and model (Qwen/Qwen2-1.5B-Instruct). To run on a single node:

python3 examples/math/gsm8k_rl.py --config examples/math/gsm8k_grpo.yaml scheduler.type=local

If you prefer to run experiments on a Ray cluster, update paths in the YAML file to point to your shared storage, and run:

python3 examples/math/gsm8k_rl.py --config examples/math/gsm8k_grpo.yaml \
  cluster.n_nodes=2 cluster.n_gpus_per_node=8 \
  cluster.fileroot=/path/to/nfs \
  scheduler.type=ray

For comprehensive setup instructions, see our quickstart guide.

For Megatron MoE router replay experiments, see the R3 Router Replay guide.

📚 Examples

Math & Reasoning

Task	Description	Performance
Math	GSM8K math reasoning with GRPO, PPO, DAPO, REINFORCE, RLOO, LitePPO, DR-GRPO, GSPO, and more	-
Multi-Turn Math	Multi-turn math agent with reward discounting across turns	Training Curve
LoRA Math	Parameter-efficient math training with LoRA (SGLang/vLLM backends)	-
Countdown	Countdown numbers game with custom rewards	Training Curve

Agentic RL

Task	Description	Performance
General Agent	General agentic training with any agentic frameworks	Guide
Hermes Online RL	End-to-end Online RL loop with Hermes agent under AReaL 2.0	Guide
Coding Agent RL	End-to-end coding-agent RL training with AReaL-SWEAgent/Claude Code Agent	Guide
Tau2 Customer Service	Customer service agent on Tau2-Bench (retail, airline, telecom)	Paper
Search Agent	End-to-end search agent with Tongyi-DeepResearch workflow	Training Curve
Tool-Integrated Reasoning	Multi-turn tool calling during reasoning (Python executor, calculator)	Training Curve
OpenAI Agents Integration	Integration with OpenAI Agents SDK for agentic workflows	-
CAMEL-AI Integration	Integration with CAMEL-AI framework for agentic RL	-

Vision-Language Models

Task	Description	Performance
VLM	Geometry3K and CLEVR Count 70K visual reasoning with GRPO	-
VLM on NPU	VLM training on Huawei NPU hardware	Benchmark Results

Alignment & Infrastructure

Task	Description	Performance
RLHF Reward Modeling	Bradley-Terry reward modeling on Anthropic HH-RLHF	Training Curve
SkyPilot Deployment	Cloud deployment with SkyPilot (GCP, AWS, Kubernetes)	Screenshots

🔧 Support Matrix

🧠 Algorithms

All RL algorithms support both asynchronous and synchronous versions by setting max_head_offpolicyness=0. See Asynchronous RL Guide.

Algorithm	Documentation	Paper	Configuration
GRPO	📖 Docs	📄 Paper	🔗 GSM8K Example
GSPO	📖 Docs	📄 Paper	🔗 GSM8K Example
PPO	📖 Docs	📄 Paper	🔗 GSM8K Example
DAPO	📖 Docs	📄 Paper	🔗 GSM8K Example
LitePPO	📖 Docs	📄 Paper	🔗 GSM8K Example
Dr.GRPO	📖 Docs	📄 Paper	🔗 GSM8K Example
REINFORCE++	-	📄 Paper	🔗 GSM8K Example
RLOO	📖 Docs	📄 Paper	🔗 GSM8K Example
SAPO	📖 Docs	📄 Paper	🔗 GSM8K Example
IcePop	📖 Docs	📄 Blog	🔗 GSM8K Example
KPop	📖 Docs	📄 Blog	🔗 GSM8K Example
M2PO	📖 Docs	📄 Paper	🔗 GSM8K Example
DPO	📖 Docs	📄 Paper	🔗 HH-RLHF Example
RLHF Reward Modeling	-	-	🔗 RLHF Example
SFT	-	-	🔗 GSM8K Example
Distillation	📖 Docs	📄 Paper	🔗 GSM8K Example

Models

Model Family	Megatron	PyTorch FSDP	PyTorch Archon	Notes
Qwen2/3	✅	✅	✅	-
Qwen3-MoE	✅	✅	✅	-
Qwen3.5	✅	✅	❌	Vision-language model
Qwen3.5-MoE	✅	✅	❌	Vision-language model
Bailing V3	✅	❌	❌	-
Qwen2.5-VL	✅	✅	❌	Vision-language model
Qwen3-VL	✅	✅	❌	Vision-language model
Gemma 3	❌	✅	❌	Vision-language model
Other Hugging Face LLM	❌	✅	❌	Compatibility depending on the version of transformers

Qwen2 and Qwen2.5 text models support RL weight updates with actor.weight_update_mode=disk in both v1 and v2. XCCL weight updates require actor._version=v1, rollout._version=v1, and actor.weight_update_mode=xccl. AWEX weight updates are not supported for these models. This restriction does not apply to Qwen2.5-VL.

Check the AI Coding Assistant Guide and Archon Reference for how to integrate new models into AReaL.

Training Backends

Backend	DP	Tensor Parallel	Sequence Parallel within TP	Context Parallel	Pipeline Parallel	Expert Parallel	1D Sequence Packing	LoRA
Megatron	✅ (ZeRO-1)	✅	✅	✅	✅	✅	✅	✅ (with vLLM inference backend)
PyTorch FSDP	✅ (FSDP2)	✅	✅	✅	❌	❌	✅	✅
PyTorch Archon	✅ (FSDP2)	✅	✅	✅	✅	✅	✅	❌

Inference Backends

Backend	Tensor Parallel	Context Parallel	Pipeline Parallel	Data Parallel Attention	Expert Parallel
vLLM	✅	❓	✅	❓	❓
SGLang	✅	❌	❌	✅	✅

📖 Resources

Tutorial

• Installation
• Quickstart
• AReaL CLI Guide
• Agentic RL
• Evaluation
• Large MoE with Megatron
• Large MoE with PyTorch Archon

Code Walkthrough

• Running GRPO on GSM8K dataset

Best Practices

• Improving Algorithm Performance
• Agent Workflow Best Practices
• Debugging
• Handling OOM Issues
• Performance Profiling

Customization

• Customize Dataset
• Customize Agentic/RVLR Rollout Workflows

Algorithms

• Asynchronous RL Explained
• PPO, GRPO, and Related Algorithms
• M2PO

Reference

• CLI Configurations
• LoRA RL
• Checkpointing
• Metrics Tracking
• Allocation Mode
• Rollout Workflow
• Agent Workflow
• AI-Assisted Development

🤝 Contributing

We warmly welcome contributions from the community! Whether you're fixing bugs, adding features, improving documentation, or helping others, your contribution is valued. Please check our Contributing Guide for detailed information.

# Fork and clone the repository
git clone https://github.com/YOUR-USERNAME/AReaL
cd AReaL

# Install uv and sync dependencies
pip install uv
# Install flash-attn pre-built wheel to avoid compiling from source
uv pip install "https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.7.16/flash_attn-2.8.3+cu128torch2.9-cp312-cp312-linux_x86_64.whl"
# Use `--extra cuda` on Linux with CUDA (installs training packages + SGLang)
uv sync --extra cuda --group dev
# For vLLM instead:
# cp pyproject.vllm.toml pyproject.toml && cp uv.vllm.lock uv.lock && uv sync --extra cuda --group dev
# Or without CUDA support
# uv sync --group dev

# Set up pre-commit hooks (formatting, linting, commit message checks)
pre-commit install --install-hooks

# Make changes
git checkout -b feat/gpt-o5
git add .
# `git commit` will automatically check your files and commit messages
git commit -m "feat: implement gpt-o5 training loop"
git push

🗺️ Future Roadmap

• Full Roadmap
• 2026 H2 Roadmap
• 2026 Q2 Roadmap

AReaL is under active development with planned minor releases weekly and major releases monthly. We warmly welcome community engagement and contributions. We are also actively hiring interns and full-time employees with open positions in both the US and China.

📢 Community

We warmly welcome everyone to join our community! We have a WeChat group for real-time discussions and a GitHub discussion board for asynchronous conversations.

Check out our Community Repository for more details and meeting materials.

🙏 Acknowledgments

We gratefully acknowledge that major contributors are from the AReaL Team at the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University and Ant Group.

We have also received invaluable assistance from the following groups (listed alphabetically):

• The Data Intelligence Lab at Ant Research for their data support

• @HwVanICI for support on vLLM, LoRA, NPU integration, and more

• The Relaxed System Lab at HKUST for seamless collaboration on numerous system-related aspects

• The SGLang team for supporting custom weight update features and their contributions during AReaL-lite development

• The Super Computing Technology (SCT) team at Ant Group for their expertise in large-scale cluster operations and maintenance

• Special thanks to @Lyken17 for providing valuable suggestions throughout the API design process

We also deeply appreciate all pioneering work from the community, particularly the ReaLHF project from OpenPsi Inc. and other outstanding projects, including but not limited to DeepScaleR, Open-Reasoner-Zero, OpenRLHF, VeRL, SGLang, QwQ, Light-R1, and DAPO.

📜 License

This project is licensed under the Apache License 2.0.

📄 Citation

@inproceedings{mei2025real,
  author       = {Mei, Zhiyu and Fu, Wei and Li, Kaiwei and Wang, Guangju and Zhang, Huanchen and Wu, Yi},
  title        = {ReaL: Efficient RLHF Training of Large Language Models with Parameter Reallocation},
  booktitle    = {Proceedings of the Eighth Conference on Machine Learning and Systems,
                  MLSys 2025, Santa Clara, CA, USA, May 12-15, 2025},
  publisher    = {mlsys.org},
  year         = {2025},
}

@misc{fu2025areal,
      title={AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning},
      author={Wei Fu and Jiaxuan Gao and Xujie Shen and Chen Zhu and Zhiyu Mei and Chuyi He and Shusheng Xu and Guo Wei and Jun Mei and Jiashu Wang and Tongkai Yang and Binhang Yuan and Yi Wu},
      year={2025},
      eprint={2505.24298},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2505.24298},
}