GPUSPECINDEX

NVIDIA · Ada Lovelace · 2022

L40 GPU accelerator

L40 GPU accelerator: 48 GB GDDR6, 300 W, 18,176 CUDA cores. Launched October 13, 2022.

Memory
48 GB
Power
300 W
Bus width
384 bit
Memory48 GB7 of 32 in Data Center
Power300 W12 of 40 by higher draw in Data Center
CUDA cores18,1763 of 47 in Data Center

Affiliate links — we may earn a commission.

Memory

Capacity
48 GB
Memory type
GDDR6
Bus width
384 bit
Bandwidth
864 GB/s

Compute

CUDA cores
18,176
SM count
142
TMUs
568
ROPs
192
FP32
90,520 GFLOPS
FP64
1,414 GFLOPS

Clocks

Base clock
735 MHz
Boost clock
2,490 MHz

Board

TDP
300 W
Form factor
datacenter

Physical

Architecture
Ada Lovelace
Series
Data Center
Launch date
October 13, 2022

Not recorded by the source: generation, codename, process and transistors, or 7 further fields.

Silicon, made visible

Die area is drawn against the largest recorded die in the catalog. Shader dots use a labeled, card-specific ratio so both early and modern GPUs remain legible.

Die size not recorded · 18,176 shadersDie size not recorded; shader grid shown without scale. Each dot = 128 shaders.18,176 shadersEach dot = 128 shadersDie size not recorded; shader grid shown without scale

Within its series, the L40 GPU accelerator ranks first of 36 for FP32 throughput and 7th of 32 for memory capacity. Its 300 W TDP is above the 75 W PCIe slot-power ceiling.

Source: Wikipedia, revision 1369488585 · catalog loaded 2026-08-16