Lambda, Stanford, and TAMU open-sourced AgentFlow, a trainable agent system utilizing Flow-GRPO turn-by-turn credit assignment to enable 7B models to beat GPT-4o on agentic and scientific benchmark suites.

Key Takeaways

  • βœ“Proves specialized multi-role loops and tool integration enable 7B models to exceed frontier API accuracy;
  • βœ“Flow-GRPO assigns reinforcement learning credit turn-by-turn rather than treating trajectories as opaque units;
  • βœ“Paper, reproducible code, and training pipelines released openly for cost-effective private agent deployment.