# AReaL

> Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.

- **URL**: https://www.freshcrate.ai/projects/AReaL
- **Author**: inclusionAI
- **Category**: AI Agents
- **Latest version**: `v1.0.4` (2026-05-07)
- **License**: Apache-2.0
- **Source**: https://github.com/inclusionAI/AReaL
- **Homepage**: https://inclusionai.github.io/AReaL/
- **Language**: Python
- **GitHub**: 5,075 stars, 473 forks
- **Registry**: github
- **Tags**: `agent`, `llm`, `llm-agent`, `llm-reasoning`, `machine-learning-systems`, `mlsys`, `python`, `reinforcement-learning`, `rl`

## Description

Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.

## Recent releases

| Version | Date | Urgency | Changes |
| --- | --- | --- | --- |
| `v1.0.4` | 2026-05-07 | High | ## What's Changed * fix: FSDP initialization for set-valued wrap class names by @Wangxiaoxiaoa in https://github.com/inclusionAI/AReaL/pull/1187 * chore: move figures into assets/figures/ and add community meeting folder by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1192 * feat(engine): lora support for MoE models (single node/ cross node) by @gursimar in https://github.com/inclusionAI/AReaL/pull/1159 * fix: handle integer device ids in ray rpc server by @Wangxiaoxiaoa in htt |
| `v1.0.3` | 2026-04-16 | High | ## What's Changed * chore(docker): add openclaw, ironclaw, zeroclaw, and nanobot-ai to runtime image by @garrett4wade in https://github.com/inclusionAI/AReaL/pull/1051 * feat(agent-service): add Agent Service microservice infrastructure by @CormickKneey in https://github.com/inclusionAI/AReaL/pull/1048 * feat(gateway): Add rollout gateway infrastructure with controller, router, and data proxy by @nuzant in https://github.com/inclusionAI/AReaL/pull/1043 * feat: estimators for kl divergence by |
| `v1.0.2` | 2026-03-17 | Medium | ## Release Note  **A massive thank you to our newest contributors who joined us for this release! The strength of this project lies in the collective expertise of the open-source community, and your work is what moves us forward.**  ### 🚀 Model & Architecture Updates + Qwen3.5 Support: Added support for both dense and MoE (Mixture-of-Experts) variants of Qwen3.5 (archon backend, DP-only).  + On-Policy Distillation: Introduced native support for on-policy distillation.  + Added opt-in s |
| `v1.0.1` | 2026-03-04 | Low | ## Release Note  A patch release that fixes a dependency issue in the docker image and enriches the documentation and testing of the OpenClaw example.  ## What's Changed * fix(config): Fix openclaw config typo and increase max_tokens_per_mb by @fishcrap in https://github.com/inclusionAI/AReaL/pull/959 * docs(openclaw): Replace hardcoded admin key with placeholder in README by @fishcrap in https://github.com/inclusionAI/AReaL/pull/967 * feat: Fully Support MIS/TIS to stablizing rollout-tra |
| `v1.0.0` | 2026-03-02 | Low | ## 🚀 Key Highlights ## Release Notes  ### **Online RL Training** - Seamlessly train any agents by configuring a `base_url` and `api_key`—no code changes required and no heavy dependencies. - Check out the [OpenClaw RL training example](https://github.com/inclusionAI/AReaL/tree/main/examples/openclaw) for more details.  ### **Archon Engine** - A fully working, PyTorch-native 5D parallel training engine. - Includes features like:   - **Automatic HF format conversion**   - **Zero-bubble |
| `v1.0.0.rc1` | 2026-02-06 | Low | Pre-release for 1.0.0. |
| `v0.5.3` | 2026-01-31 | Low | ## Highlights  This is a patch release primarily for delivering the latest docker image for testing.  We will include well-documented features in the next major release. |
| `v0.5.2` | 2026-01-26 | Low | ## Highlights  This is a patch release primarily for delivering the latest docker image with torch 2.9.1, vllm 0.14.0, and sglang 0.5.7 supports.  We will include well-documented features in the next major release. |
| `v0.5.1` | 2025-12-18 | Low | ## Highlights  This is a patched release upon v0.5.0.  + A new docker image with `math-verify` and the latest `ruff`. + Support for PPO critic model support with Megatron engine. + Refactored FSDP/Megatron engine implementations. + Implement efficient RPC tensor transfer with `RTensor` (aka the original `DistributedBatch`). + Beam seach support for vLLM.  ## What's Changed * fix: change checkpoint cleanup flag to fix update_weights_from_disk in single-controller mode by @HwVanICI in h |
| `v0.5.0` | 2025-12-10 | Low | ## Highlights  The newly released v0.5.0 of AReaL introduces two core innovations: Seamless Agentic RL and the Single Controller architecture:  + **Seamless Agentic RL**: AReaL provides a seamless intelligent agent training service via OpenAI-compatible APIs. This facilitates seamless collaboration among environment providers, algorithm developers, and system engineers, forming a zero-friction pipeline in complex engineering workflows and significantly boosting development efficiency and sys |

## Citation

- HTML: https://www.freshcrate.ai/projects/AReaL
- Markdown: https://www.freshcrate.ai/projects/AReaL.md
- Dependencies JSON: https://www.freshcrate.ai/api/projects/AReaL/deps

_Generated by freshcrate.ai. Indexes github releases for AI-agent ecosystem packages._