view post Post 161 AsyncGRPOTrainer is now stable in TRL v1.18.0!generation and training run at the same time on separate GPUs, instead of taking turnswe wrote up how to run it, what you can train with it and how it compares with GRPOTrainer: trl-lib/async-grpo-stable See translation Reply
Math Environments Collection Math RL environments graded exactly, by Lean, or by a rubric judge; and Qwen3.5-2B trained on olympiad proofs. β’ 7 items β’ Updated 7 days ago
Math Environments Collection Math RL environments graded exactly, by Lean, or by a rubric judge; and Qwen3.5-2B trained on olympiad proofs. β’ 7 items β’ Updated 7 days ago
Math Environments Collection Math RL environments graded exactly, by Lean, or by a rubric judge; and Qwen3.5-2B trained on olympiad proofs. β’ 7 items β’ Updated 7 days ago