DEPLOYDatabase

Chip comparison

NVIDIA T4 Tensor Core GPU 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.

FieldNVIDIA T4 Tensor Core GPUSambaNova SN40L
Process nodeTSMC 12FFNTSMC N5
Transistors (B)13.6
Die size545 mm²
TDP70 W700 W
Memory16 GB GDDR61500 GB DDR5 (three-tier: 64 MB SRAM + 64 GB HBM3 + 1.5 TB DDR5)
Memory bandwidth320 GB/s
FP16 (dense)65 TFLOPS638 TFLOPS
FP8 (dense)1,250 TFLOPS
INT8 (dense)130 TOPS
Form factorPCIe (single-slot, low-profile)PCIe (SN40L DataScale system)
Announced2018-09-132023-09-19
Released2019-01-012024-01-01

Robots running NVIDIA T4 Tensor Core GPU

No robots publicly running it yet.

Robots running SambaNova SN40L

No robots publicly running it yet.

Data centers with NVIDIA T4 Tensor Core GPU

No data centers publicly running it yet.

Data centers with SambaNova SN40L

No data centers publicly running it yet.

Common questions

NVIDIA T4 Tensor Core GPU vs SambaNova SN40L: which is faster for training?

SambaNova SN40L has 19.23x the dense FP16/BF16 throughput of the other (NVIDIA T4 Tensor Core GPU: 65 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.

NVIDIA T4 Tensor Core GPU vs SambaNova SN40L: which has more memory?

SambaNova SN40L carries more HBM (NVIDIA T4 Tensor Core GPU: 16 GB GDDR6; 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.

NVIDIA T4 Tensor Core GPU vs SambaNova SN40L: which is more power-efficient?

Dense FP16/BF16 TFLOPS per watt: NVIDIA T4 Tensor Core GPU 0.93 (65 ÷ 70 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.

See also: every chip comparison · NVIDIA T4 Tensor Core GPU full page · SambaNova SN40L full page.