🛰️ Daily AI Frontier
‹ back to 2026-07-17

LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget

Research Efficiency & Systems

Merged summary

TL;DR - LongStraw is an architecture-aware execution stack that makes million-token RL post-training (via GRPO) feasible on a fixed GPU budget, closing the gap between long-context inference (~1M tokens) and post-training that typically stays at ≤256K tokens—especially relevant for long-trajectory AI agents.

  • Reduces the live training graph by evaluating the shared prompt without autograd, retaining only model state needed by later tokens, and replaying short response branches one at a time (trading extra replay time for memory).
  • On 8 H20 GPUs, it completes grouped Qwen3.6-27B scoring and response backward at 2.1M positions (groups of 2 and 8); larger group size adds only 0.21 GB peak memory, and a stress test reaches 4.46M positions.
  • On 32 H20 GPUs, it validates the end-to-end path for a 2.1M-token prompt across all 78 layers of the compressed-attention MoE GLM-5.2.
  • Authors explicitly frame results as demonstrating execution capacity, not training correctness—prompt state is detached and some distributed forward/gradient composition paths remain incomplete.

Sources (1)

LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget

arXiv cs.LG Changhai Zhou, Kieran Liu, Yuhua Zhou, Qian Qiao, Jun Gao, Harry Zhang, Irvine Lu, Nolan Ho, Lucian Li, Andrew Lei, Cleon Cheng, Steven Chiang, Yihang Zeng, Di Zhang, Rio Yang, Kaijie Chen, Andrew Chen, Pony Ma, Weizhong Zhang, Cheng Jin 2026-07-16 arXiv:2607.14952

TL;DR - LongStraw is an architecture-aware execution stack that makes million-token RL post-training (via GRPO) feasible on a fixed GPU budget, closing the gap between long-context inference (~1M tokens) and post-training that typically stays at ≤256K tokens—especially relevant for long-trajectory AI agents.

  • Reduces the live training graph by evaluating the shared prompt without autograd, retaining only model state needed by later tokens, and replaying short response branches one at a time (trading extra replay time for memory).
  • On 8 H20 GPUs, it completes grouped Qwen3.6-27B scoring and response backward at 2.1M positions (groups of 2 and 8); larger group size adds only 0.21 GB peak memory, and a stress test reaches 4.46M positions.
  • On 32 H20 GPUs, it validates the end-to-end path for a 2.1M-token prompt across all 78 layers of the compressed-attention MoE GLM-5.2.
  • Authors explicitly frame results as demonstrating execution capacity, not training correctness—prompt state is detached and some distributed forward/gradient composition paths remain incomplete.
item →