Google TPU Ironwood
v7 TPU pods scaling to 9,216 chips, available through Google Cloud and used for Gemini.
Full Google TPU Ironwood record →Head to head
Google TPU Ironwood vs Intel Gaudi 3
Google TPU Ironwood (Google) and Intel Gaudi 3 (Intel) both sit in Chips & Compute. Google TPU Ironwood shipped more tracked changes in the last 30 days (16 vs 7). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | Google TPU Ironwood | Intel Gaudi 3 Intel |
|---|---|---|
| Segment | Chips & Compute | Chips & Compute |
| Version | Ironwood (7th generation) | 1.24.1 |
| Entry cost | from $5.4/unit | Hardware/cloud accelerator, pricing not publicly listed |
| Pricing tiers | On Demand (us-central1) $12 · 1-year Commitment (us-central1) $8.4 · 3-year Commitment (us-central1) $5.4 · DWS Flex-start (us-central1) $6 · DWS Calendar Mode (us-central1) $8.4 · On Demand (europe-west2) $13.2 · 1-year Commitment (europe-west2) $9.24 · 3-year Commitment (europe-west2) $5.94 | OEM/cloud free |
| Model stack | Gemini · Ironwood · Ironwood (TPU v7) · Ironwood TPU · Ironwood TPU (7th Generation) · TPU v7 · TPU7x · TPU7x (Ironwood) · XLA compiler | Accelerator · Intel Gaudi 2 · Intel Gaudi 3 |
| Context window | — | n/a |
| Public API | ||
| Routes models | ||
| Changes / 30d | 16 | 7 |
| Origin | Global | Global |
| Capabilities | 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 | 128GB HBM2e, Open Ethernet scale-out, PyTorch native, oneAPI toolchain, Price-performance focus, PyTorch Support, vLLM Integration, DeepSpeed Training |
Intel Gaudi 3
Gaudi 3 targets price-performance against Nvidia parts with open Ethernet scale-out and an oneAPI/PyTorch software path.
Full Intel Gaudi 3 record →