AMD Instinct MI355X
288GB HBM3E accelerators with the ROCm software stack, used by several inference providers.
Full AMD Instinct MI355X record →Head to head
AMD Instinct MI355X vs Google TPU Ironwood
AMD Instinct MI355X (AMD) and Google TPU Ironwood (Google) both sit in Chips & Compute. AMD Instinct MI355X is the cheaper entry point at Enterprise hardware, pricing not published. Google TPU Ironwood shipped more tracked changes in the last 30 days (11 vs 3). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | AMD Instinct MI355X AMD | Google TPU Ironwood |
|---|---|---|
| Segment | Chips & Compute | Chips & Compute |
| Version | MI355X | 7th-generation (Ironwood) |
| Entry cost | Enterprise hardware, pricing not published | from $5.4/unit |
| Pricing tiers | Open Source free | On Demand (us-central1) $12 · DWS Flex-start (us-central1) $6 · DWS Calendar Mode (us-central1) $8.4 · 1-year Commitment (us-central1) $8.4 · 3-year Commitment (us-central1) $5.4 · On Demand (europe-west2) $13.2 · DWS Flex-start (europe-west2) $6 · DWS Calendar Mode (europe-west2) $8.4 |
| Model stack | CDNA · CDNA 4 · ROCm | Gemini · Google TPU Ironwood · Ironwood · Ironwood (TPU v7) · Ironwood TPU · Ironwood TPU (7th Generation) · TPU v7 · TPU7x · TPU7x (Ironwood) · XLA compiler |
| Context window | — | — |
| Public API | ||
| Routes models | ||
| Changes / 30d | 3 | 11 |
| Origin | Global | Global |
| Capabilities | 288GB HBM3E, FP4/FP6 support, ROCm 7, open rack designs, DPX compute partition mode, CPX compute partition mode, NPS2 memory partitioning, GPU virtualization support | 9,216-chip pods, inference optimised, liquid cooled, JAX + PyTorch/XLA, Large-scale training, Reasoning and inference, 9,216 liquid-cooled chips per pod, 42.5 ExaFlops performance |
Google TPU Ironwood
v7 TPU pods scaling to 9,216 chips, available through Google Cloud and used for Gemini.
Full Google TPU Ironwood record →