Toolsslug: maxtext

MaxText

Google's open-source high-performance LLM training framework for scalable training on TPUs and GPUs.

Curated Intelligence

1 items · Page 1/1
vLLM 官方博客T1·Infra
78 pts
Google TPU v7 Cuts Small Dense Model RL Cycle 107x via vLLM

Google's ASA and Ranking AI Research teams optimized the open-source TPU stack (vLLM, Tunix, MaxText, Pallas) for RL training of a 0.6B dense model on TPU v7, cutting the full rollout-to-update cycle from roughly 100 minutes to about 56 seconds (~183K tokens/s), a 107x speedup while retaining the full 8,192-token response budget. All changes were upstreamed, so developers can reproduce the path via the merged tpu-recipes recipe without redoing bring-up work.