HBM3E per Ironwood TPU v7 (inference-era ASIC); 9,216-chip pods, 42.5 FP8 ExaFLOPS, 4,614 FP8 TFLOPS/chip

Name: HBM3E per Ironwood TPU v7 (inference-era ASIC); 9,216-chip pods, 42.5 FP8 ExaFLOPS, 4,614 FP8 TFLOPS/chip
License: https://aidatacenterguide.com/terms#intellectual-property

192 GB / 7.4 TB/sobserved

Value kind	observed — Reported measurements, counts, and specifications keep the precision and scope stated by their source; an exact specification is not treated as a range.
Scope	Decimal GB, as Google's Ironwood announcement states; the same figure is carried at its canonical home in Chapter 7.4 (claim hbm3e-per-ironwood-tpu-v7-chip-7-37-tb-s-4-614, 192 GB at 7.37 TB/s). 192 GiB would be ~206 GB and is not what the cited source says.
As of	2025
Source	Google Cloud
Review	checking…review by 2026-10-06 · standard cadence
Recorded changes	last 2026-09-16 · 3 revisions tracked
Claim id	hbm3e-per-ironwood-tpu-v7-inference-era-asic-9

Where the guide uses it

1.3 Inference Data Centers: Bursty, Distributed, Always-On

← Full numbers register — every date-stamped figure in the guide, with revision history.