한국어EN
Inference Infrastructure & ServingGoogle Cloud

Google Cloud makes TPU7x generally available (GA)

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

TPU7x generally available (GA) — 192GB HBM per chip, 7.37TB/s

Why it matters

Ironwood went from announcement to actual supply

Source — Primary sources first; secondary reports are labelled “Press”

Officialdocs.cloud.google.comhttps://docs.cloud.google.com/tpu/docs/tpu7x

Added in Week 35, 2026 · See that week

On the map — ⑧ Inference Infrastructure line

  1. Previous stopIronwood
  2. This recordGoogle Cloud makes TPU7x generally available (GA)
  3. Next stopGoogle Cloud previews its 8th-generation TPUs at Cloud Next
See it on the map

Same week

Self-supervised multi-resolution learning of thin-film SEM defect morphologyPaperMCP Dev Summit North America 2026 (New York, 1,200 attendees)OfficialHarness engineering for coding agent usersOfficialGemma 4Official

Nearby on the same axis

vLLM introduces Model Runner V2OfficialHugging Face archives the text-generation-inference repositoryOfficialGoogle Cloud previews its 8th-generation TPUs at Cloud NextPressFlashAttention-4Paper