PyyanThe AI Index

Pyyan / Compare / MI400 vs B300 Blackwell Ultra vs B200 vs H100 vs TPU v7 Ironwood

MI400 vs B300 Blackwell Ultra vs B200 vs H100 vs TPU v7 Ironwood

5 of 5

GPU & AI Chips · verified 13 Aug 2026

×MI400AMDcurrent
×B300 Blackwell UltraNVIDIAcurrent
×B200NVIDIAcurrent
×H100NVIDIAcurrent
×TPU v7 IronwoodGooglecurrent
SpecificationMI400B300 Blackwell UltraB200H100TPU v7 Ironwood
SummaryDouble NVIDIA's memory capacity.15 PFLOPS dense FP4.The current volume Blackwell part.The workhorse of the previous generation.Inference-oriented, Google Cloud only.
Memory432GB HBM4288GB HBM3e192GB HBM3e80GB HBM3192GB HBM3e
Bandwidth19.6 TB/s8 TB/s——7.37 TB/s
Price—~$40–50K (verify)$30–50K$25–40K—
VendorAMDNVIDIANVIDIANVIDIAGoogle
CategoryGPU & AI ChipsGPU & AI ChipsGPU & AI ChipsGPU & AI ChipsGPU & AI Chips
OfficialAMD ↗NVIDIA ↗NVIDIA ↗NVIDIA ↗Google ↗

Highlighted rows are where these differ.

MI400

  • 19.6 TB/s bandwidth
  • 40 PFLOPS FP4
  • $7.2B in commitments

Best for large-model inference.

Full spec sheet →

B300 Blackwell Ultra

  • 8 TB/s memory bandwidth
  • Per-GPU pricing widely misquoted — one source cites rack pricing

Best for frontier training.

Full spec sheet →

B200

  • Sold out through mid-2026
  • ~3.6M unit backlog

Best for general training.

Full spec sheet →

H100

The workhorse of the previous generation.

Best for value.

Full spec sheet →

TPU v7 Ironwood

  • Requires JAX
  • Not purchasable or on any marketplace

Best for GCP inference.

Full spec sheet →