AReaL
Large-scale asynchronous reinforcement learning system for training reasoning and agentic LLMs - simple, flexible, and production-ready.
Overview
Requirements
Recommended VPS
Hostinger · KVM 4
4 vCPU · 16384 MB · 200 GB
Hostinger · KVM 4
4 vCPU · 16384 MB · 200 GB
Hostinger · KVM 8
8 vCPU · 32256 MB · 400 GB
Affiliate disclosure
Docker Compose
# Generated by Run This Ai — docker-compose.yml
services:
areal:
image: ghcr.io/areal-project/areal:latest
restart: unless-stopped
ports:
- 8080:8080
volumes:
- ./data/areal:/data
Related tools
AutoGPT
Build, Deploy, and Run AI Agents — autonomous AI that chains tasks to reach goals
Browser Use
Make websites accessible for AI agents — automate any online task
n8n
Workflow automation with AI nodes
OpenHands
AI-driven development platform — an autonomous coding agent that writes, runs, and debugs code
MetaGPT
Multi-Agent Framework for AI Software Teams
Dify
Open-source LLM app development platform
Guides & articles
AReaL Tutorial: Train Your First AI Agent with Asynchronous Reinforcement Learning
A hands-on guide to training agents with AReaL: pull the runtime, point it at any agent app, and launch async RL training.
AReaL: Large-Scale Asynchronous Reinforcement Learning for Reasoning and Agentic LLMs
Meet AReaL, the open-source asynchronous RL system from Tsinghua IIIS and Ant Group for training reasoning and agentic LLMs at scale.