GEPA: New AI Algorithm Outperforms Reinforcement Learning, Uses 35x Fewer Rollouts for Complex LLM Systems

GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning

Authors: Lakshya A Agrawal, Shangyin Tan, Dilara Soylu, Noah Ziems, Rishi Khare, Krista Opsahl-Ong, Arnav Singhvi, Herumb Shandilya, Michael J Ryan, Meng Jiang, Christopher Potts, Koushik Sen, Alexandros G. Dimakis, Ion Stoica, Dan Klein, Matei Zaharia, Omar KhattabPaper: https://arxiv.org/abs/2507.19457What was done? The authors introduced GEPA (Genetic-Pareto), a novel algorithm for optimizing prompts in complex, multi-module AI systems. Instead of relying on traditional reinforcement learning...