GitHub repository
patchy631/ART
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen2.5, Qwen3, Llama, and more!
Repository metrics
- Stars · Nov 15, 2026
- 3
- Leaderboard status
- Outside the current top 1,000
Language Python
Source: GitHubLB snapshots.