DEPLOYDatabase

Chip comparison

AWS Inferentia 2 vs Google TPU v7 (Ironwood)

Side-by-side specs

Straight from each vendor's datasheet. Winner in bold: higher throughput, memory, or transistor count; lower power draw or price. Sparse (2:4) throughput roughly doubles the FP8/INT8 numbers shown.

FieldAWS Inferentia 2Google TPU v7 (Ironwood)
Process nodeTSMC N7
TDP250 W700 W
Memory32 GB HBM3192 GB HBM3e
Memory bandwidth820 GB/s7,400 GB/s
BF16 (dense)190 TFLOPS4,614 TFLOPS
FP8 (dense)380 TFLOPS9,228 TFLOPS
Form factorInf2 EC2 instanceOAM (Ironwood pod, up to 9,216 chips)
Announced2022-12-012025-04-09
Released2023-04-012025-12-01

Robots running AWS Inferentia 2

No robots publicly running it yet.

Robots running Google TPU v7 (Ironwood)

No robots publicly running it yet.

Data centers with AWS Inferentia 2

No data centers publicly running it yet.

Data centers with Google TPU v7 (Ironwood)

No data centers publicly running it yet.

Common questions

AWS Inferentia 2 vs Google TPU v7 (Ironwood): which is faster for training?

Google TPU v7 (Ironwood) has 24.28x the dense FP8 throughput of the other (AWS Inferentia 2: 380 TFLOPS; Google TPU v7 (Ironwood): 9,228 TFLOPS). Real-world training speed also depends on memory bandwidth, interconnect, and software stack: see the spec diff for those inputs.

AWS Inferentia 2 vs Google TPU v7 (Ironwood): which has more memory?

Google TPU v7 (Ironwood) carries more HBM (AWS Inferentia 2: 32 GB HBM3; Google TPU v7 (Ironwood): 192 GB HBM3e). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.

AWS Inferentia 2 vs Google TPU v7 (Ironwood): which is more power-efficient?

Dense FP8 TFLOPS per watt: AWS Inferentia 2 1.52 (380 ÷ 250 W); Google TPU v7 (Ironwood) 13.18 (9,228 ÷ 700 W). Google TPU v7 (Ironwood) wins at the die level; system-level efficiency also depends on cooling and interconnect.

See also: every chip comparison · AWS Inferentia 2 full page · Google TPU v7 (Ironwood) full page.