Python · Apache-2.0
OpenPipe/ART
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!
★ 10.8k
Stars
1k forks · 185 open issues
+2.4/day
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
41.7
Momentum / 100
see the breakdown below
20mo ago
Created
last push 2d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 10.2 | +2.4/day, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 4.7 | last week ran 1.32× the monthly pace |
| Freshness | 15 | 15.0 | last push 2d ago; decays to zero at 180 days |
| Scale | 15 | 11.7 | 10.8k stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 41.7 | sum of the above |
Detail
| Topics | agentagentic-aigrpollmsloraqwenqwen3reinforcement-learningrl |
| Matched on | AI Agents (agentic-ai) |
| Short window | +17 stars over 7 days = 2.4/day |
| Long window | +48 stars over 26 days = 1.8/day |
| Lifetime average | 17.93 stars/day over 605 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |