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 Huawei Ascend 910C
Google TPU Ironwood (Google) and Huawei Ascend 910C (Huawei) both sit in Chips & Compute. Google TPU Ironwood shipped more tracked changes in the last 30 days (10 vs 4). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | Google TPU Ironwood | Huawei Ascend 910C Huawei |
|---|---|---|
| Segment | Chips & Compute | Chips & Compute |
| Version | Ironwood (7th generation) | Ascend 910C |
| Entry cost | from $5.4/unit | Pricing is not publicly available (enterprise hardware) |
| Pricing tiers | On-Demand (us-central1 Iowa) $12 · DWS Flex-start price (us-central1 Iowa) $6 · DWS Calendar Mode price (us-central1 Iowa) $8.4 · 1-year Commitment (us-central1 Iowa) $8.4 · 3-year Commitment (us-central1 Iowa) $5.4 · On-Demand (europe-west2 London) $13.2 · DWS Flex-start price (europe-west2 London) $6 · DWS Calendar Mode price (europe-west2 London) $8.4 | — |
| Model stack | Gemini · Ironwood · Ironwood (TPU v7) · Ironwood TPU · Ironwood TPU (7th Generation) · TPU v7 · TPU7x · TPU7x (Ironwood) · XLA compiler | Ascend 910C · CANN / MindSpore |
| Context window | — | — |
| Public API | ||
| Routes models | ||
| Changes / 30d | 10 | 4 |
| Origin | Global | China |
| 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 | dual-die 910C, CloudMatrix 384 racks, CANN toolkit, MindSpore, CANN 异构计算架构, MindSpore AI框架, MindStudio 全流程工具链, NPU |
Huawei Ascend 910C
Domestic training and inference silicon with the CANN/MindSpore stack and CloudMatrix racks.
Full Huawei Ascend 910C record →