Google open-sources TPU codesign guide: 8 principles for LLM/MoE performance

rseroter · x · 2026-11-19

Google launched the open-source AI-Hypercomputer/tpu-codesign guide on its developers blog, covering how to design and optimize models for LLMs, MoE, and other compute-heavy workloads on TPUs (targeting Ironwood/TPU v7x and beyond). Key thesis: architectural choices made before the first training step set permanent ceilings on MFU, cost, and serving latency. The guide groups rules into three tiers:

New model designs should start with shape rules; adapting existing models mainly needs the parallelism section.

Original post →

More from Infra

Infra channel →