NVIDIA open-sources FlashREINFORCE: agentic RL that beats GRPO with half the rollouts
Around September 13–14, 2026, NVIDIA released FlashREINFORCE (Apache 2.0, Molt integration): agentic RL that, per TechTimes, beats GRPO with half the rollouts, posting 38.0 vs 36.3 on Qwen2.5-Math-1.5B.