DEPLOYDatabase

Chip comparison

Intel Gaudi 2 vs NVIDIA L40S

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.

FieldIntel Gaudi 2NVIDIA L40S
Process nodeTSMC N7TSMC 4N
Transistors (B)76.3
TDP600 W350 W
Memory96 GB HBM2e48 GB GDDR6
Memory bandwidth2,450 GB/s864 GB/s
FP16 (dense)362 TFLOPS
BF16 (dense)432 TFLOPS362 TFLOPS
FP8 (dense)733 TFLOPS
INT8 (dense)733 TOPS
Form factorOAMPCIe
Announced2022-05-102023-08-08
Released2022-09-01

Robots running Intel Gaudi 2

No robots publicly running it yet.

Robots running NVIDIA L40S

No robots publicly running it yet.

Data centers with Intel Gaudi 2

No data centers publicly running it yet.

Data centers with NVIDIA L40S

No data centers publicly running it yet.

Common questions

Intel Gaudi 2 vs NVIDIA L40S: which is faster for training?

NVIDIA L40S has 1.70x the dense FP16/BF16 throughput of the other (Intel Gaudi 2: 432 TFLOPS; NVIDIA L40S: 733 TFLOPS). Real-world training speed also depends on memory bandwidth, interconnect, and software stack: see the spec diff for those inputs.

Intel Gaudi 2 vs NVIDIA L40S: which has more memory?

Intel Gaudi 2 carries more HBM (Intel Gaudi 2: 96 GB HBM2e; NVIDIA L40S: 48 GB GDDR6). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.

Intel Gaudi 2 vs NVIDIA L40S: which is more power-efficient?

Dense FP16/BF16 TFLOPS per watt: Intel Gaudi 2 0.72 (432 ÷ 600 W); NVIDIA L40S 2.09 (733 ÷ 350 W). NVIDIA L40S wins at the die level; system-level efficiency also depends on cooling and interconnect.

See also: every chip comparison · Intel Gaudi 2 full page · NVIDIA L40S full page.