NVIDIA L40 Professional Graphics Card

Product status: Official | Last Update: 2023-08-09 | Report Error

Graphics Processor

AD102-895/895A

Cores : TMUs : ROPs

NVIDIA L40

Manufacturer

NVIDIA

Original Series

Data Center Ada

S/N Code

699-2G133-0250-XXX

Model

NVIDIA PG133 SKU 250

GPU Model

AD102-895/895A (AD102)

Architecture

Ada

Fabrication Process

4 nm (TSMC 4N)

Die Size

608 mm²

Transistors Count

76.3B

Transistors Density

125.5M ^TRAN/_mm²

CUDAs

18176

Tensor Cores

568

Ray Tracing Cores

142

TMUs

568

ROPs

192

Base Clock

TBC MHz

Boost Clock

2490 MHz

Memory Clock

TBC MHz

Effective Memory Clock

TBC Mbps

Memory Size

49152 MB

Memory Type

GDDR6

Memory Bus Width

384-bit

Memory Bandwidth

0.0 GB/s

Interface

PCI-Express 4.0 x16

Height

2-slot

Power Connectors

1× 16-pin

TDP/TBP

300 W

Recommended PSU

600 W

Multi-GPU Support

2-way

DisplayPort

4 × (DP 1.4a)

DirectX

12.2

Vulkan

1.2

OpenGL

4.6

OpenCL

3.0

Pixel Fillrate

478.1 GPixels/s

Texture Fillrate

1.4 TTexel/s

Peak FP32

90.5 TFLOPS

FP32 Perf. per Watt

301.7 GFLOPS/W

FP32 Perf. per mm²

148.9 GFLOPS/mm²

Pictures

Data Center Ada Series2

	Model	Cores	Boost Clock	Memory Clock	Memory Config.
	NVIDIA L40S	18176	2520 MHz	-	48 GB G6 384b
	NVIDIA L40	18176	2490 MHz	-	48 GB G6 384b

AD102 GPU Cards7

Model	Cores	Boost Clock	Memory Clock	Memory Config.
NVIDIA TITAN RTX ADA	TBC	TBC	TBC	TBC
NVIDIA L40	18176	2490 MHz	-	48 GB G6 384b
NVIDIA L40S	18176	2520 MHz	-	48 GB G6 384b
NVIDIA RTX 6000 ADA	18176	2506 MHz	20 GB/s	48 GB G6 384b
NVIDIA GeForce RTX 4090 Ti	TBC	TBC	TBC	TBC
NVIDIA GeForce RTX 4090	16384	2520 MHz	21 GB/s	24 GB G6X 384b
NVIDIA RTX 5000 ADA	12800	2598 MHz	18 GB/s	32 GB G6 256b