Chip comparison
Google TPU v5e 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 v5e | SambaNova SN40L |
|---|---|---|
| Process node | — | TSMC N5 |
| TDP | 170 W | 700 W |
| Memory | 16 GB HBM2 | 1500 GB DDR5 (three-tier: 64 MB SRAM + 64 GB HBM3 + 1.5 TB DDR5) |
| Memory bandwidth | 819 GB/s | — |
| FP16 (dense) | — | 638 TFLOPS |
| BF16 (dense) | 197 TFLOPS | — |
| FP8 (dense) | — | 1,250 TFLOPS |
| INT8 (dense) | 393 TOPS | — |
| Form factor | OAM (v5e pod, up to 256 chips) | PCIe (SN40L DataScale system) |
| Announced | 2023-08-29 | 2023-09-19 |
| Released | 2023-11-01 | 2024-01-01 |
Robots running Google TPU v5e
No robots publicly running it yet.
Robots running SambaNova SN40L
No robots publicly running it yet.
Data centers with Google TPU v5e
Data centers with SambaNova SN40L
No data centers publicly running it yet.
Common questions
Google TPU v5e vs SambaNova SN40L: which is faster for training?
SambaNova SN40L has 6.35x the dense FP16/BF16 throughput of the other (Google TPU v5e: 197 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 v5e vs SambaNova SN40L: which has more memory?
SambaNova SN40L carries more HBM (Google TPU v5e: 16 GB HBM2; 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 v5e vs SambaNova SN40L: which is more power-efficient?
Dense FP16/BF16 TFLOPS per watt: Google TPU v5e 1.16 (197 ÷ 170 W); SambaNova SN40L 1.79 (1,250 ÷ 700 W). SambaNova SN40L wins at the die level; system-level efficiency also depends on cooling and interconnect.
Google TPU v5e vs SambaNova SN40L: which is more widely used?
Google TPU v5e: 0 robots and 1 data center publicly running it. SambaNova SN40L: 0 robots and 0 data centers. Empty here means no operator has said publicly, not zero adoption in the market.
See also: every chip comparison · Google TPU v5e full page · SambaNova SN40L full page.