DEPLOYDatabase

Chip comparison

NVIDIA B100 (Blackwell) 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.

FieldNVIDIA B100 (Blackwell)NVIDIA L40S
Process nodeTSMC 4NPTSMC 4N
Transistors (B)76.3
TDP700 W350 W
Memory192 GB HBM3e48 GB GDDR6
Memory bandwidth8,000 GB/s864 GB/s
FP16 (dense)1,750 TFLOPS362 TFLOPS
BF16 (dense)1,750 TFLOPS362 TFLOPS
FP8 (dense)3,500 TFLOPS733 TFLOPS
INT8 (dense)733 TOPS
Form factorSXM6 (700 W envelope)PCIe
Announced2024-03-182023-08-08
Released2024-11-01

Robots running NVIDIA B100 (Blackwell)

No robots publicly running it yet.

Robots running NVIDIA L40S

No robots publicly running it yet.

Data centers with NVIDIA B100 (Blackwell)

No data centers publicly running it yet.

Data centers with NVIDIA L40S

No data centers publicly running it yet.

Common questions

NVIDIA B100 (Blackwell) vs NVIDIA L40S: which is faster for training?

NVIDIA B100 (Blackwell) has 4.77x the dense FP8 throughput of the other (NVIDIA B100 (Blackwell): 3,500 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.

NVIDIA B100 (Blackwell) vs NVIDIA L40S: which has more memory?

NVIDIA B100 (Blackwell) carries more HBM (NVIDIA B100 (Blackwell): 192 GB HBM3e; NVIDIA L40S: 48 GB GDDR6). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.

NVIDIA B100 (Blackwell) vs NVIDIA L40S: which is more power-efficient?

Dense FP8 TFLOPS per watt: NVIDIA B100 (Blackwell) 5.00 (3,500 ÷ 700 W); NVIDIA L40S 2.09 (733 ÷ 350 W). NVIDIA B100 (Blackwell) wins at the die level; system-level efficiency also depends on cooling and interconnect.

See also: every chip comparison · NVIDIA B100 (Blackwell) full page · NVIDIA L40S full page.