Termsslug: asyncgrpo

AsyncGRPO

异步 GRPO

Asynchronous Group Relative Policy Optimization RL algorithm in TRL for agent trajectory training.

Curated Intelligence

1 items · Page 1/1
Marktechpost (@Marktechpost)T2·Open Source
78 pts
Hugging Face Releases Qwen3.8 Coding LoRA Adapter Boosting Success to 65%

Hugging Face released Qwen3.8-27B-pi-adapted, a LoRA adapter trained for the pi coding agent that lifts SWE-bench Pro pass@1 from 39.2% to 64.7% and Terminal-Bench 2.1 from 39.5% to 55.8%. The approach pairs one small adapter per model and harness, trained with AsyncGRPO reinforcement learning inside pi and rewarded for fewer tool calls, cutting median turns on solved tasks from 59 to 23 while using roughly 4.5x fewer tokens than the base model's thinking mode. The rank-32 LoRA ships under Apache-2.0 and runs on vLLM or as a GGUF adapter in llama.cpp.