Chip comparison
Google TPU v7 (Ironwood) vs SambaNova SN40L
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.
| Field | Google TPU v7 (Ironwood) | SambaNova SN40L |
|---|---|---|
| Process node | — | TSMC N5 |
| TDP | 700 W | 700 W |
| Memory | 192 GB HBM3e | 1500 GB DDR5 (three-tier: 64 MB SRAM + 64 GB HBM3 + 1.5 TB DDR5) |
| Memory bandwidth | 7,400 GB/s | — |
| FP16 (dense) | — | 638 TFLOPS |
| BF16 (dense) | 4,614 TFLOPS | — |
| FP8 (dense) | 9,228 TFLOPS | 1,250 TFLOPS |
| Form factor | OAM (Ironwood pod, up to 9,216 chips) | PCIe (SN40L DataScale system) |
| Announced | 2025-04-09 | 2023-09-19 |
| Released | 2025-12-01 | 2024-01-01 |
Robots running Google TPU v7 (Ironwood)
No robots publicly running it yet.
Robots running SambaNova SN40L
No robots publicly running it yet.
Data centers with Google TPU v7 (Ironwood)
No data centers publicly running it yet.
Data centers with SambaNova SN40L
No data centers publicly running it yet.
Common questions
Google TPU v7 (Ironwood) vs SambaNova SN40L: which is faster for training?
Google TPU v7 (Ironwood) has 7.38x the dense FP8 throughput of the other (Google TPU v7 (Ironwood): 9,228 TFLOPS; SambaNova SN40L: 1,250 TFLOPS). Real-world training speed also depends on memory bandwidth, interconnect, and software stack: see the spec diff for those inputs.
Google TPU v7 (Ironwood) vs SambaNova SN40L: which has more memory?
SambaNova SN40L carries more HBM (Google TPU v7 (Ironwood): 192 GB HBM3e; SambaNova SN40L: 1500 GB DDR5 (three-tier: 64 MB SRAM + 64 GB HBM3 + 1.5 TB DDR5)). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.
Google TPU v7 (Ironwood) vs SambaNova SN40L: which is more power-efficient?
Dense FP8 TFLOPS per watt: Google TPU v7 (Ironwood) 13.18 (9,228 ÷ 700 W); SambaNova SN40L 1.79 (1,250 ÷ 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 · Google TPU v7 (Ironwood) full page · SambaNova SN40L full page.