Interactive 3D Model

Loading viewer...

Drag to rotate ┬╖ Scroll to zoom

ARCHITECTURE

Ada Lovelace

SERIES

L40S

Memory

48GB GDDR6

Bandwidth

864 GB/s

TDP

350W

In StockL40S┬╖Best for: Data Center Versatility

NVIDIA L40S GPU

Ada Lovelace

|Data Center GPUs

Overview

The NVIDIA L40S GPU is a versatile data center accelerator designed for AI inference, graphics rendering, and video processing. Powered by Ada Lovelace architecture with 48GB GDDR6 memory, it handles diverse workloads with ease.

Key Features

48GB GDDR6
Ada Lovelace
AV1 Encoding

Ideal Use Cases

ai inferencevirtualization

Full Specifications

Complete technical specifications for the L40S.

Memory

Memory: 48GB GDDR6
Memory Bandwidth: 864 GB/s

Compute

Tensor Cores: 568
CUDA Cores: 18176
FP8 TFLOPS: 1467
FP16 TFLOPS: 733
TF32 TFLOPS: 91
FP64 TFLOPS: —

Interconnect

Interconnect: PCIe 4.0 x16

Physical

TDP: 350W
Form Factor: PCIe Gen4
Cooling: Active Fan

Manufacturing

Launch Date: Q3 2023
Manufacturing Process: 5nm

Ready to deploy the L40S?

Get a personalized quote within 24 hours. Our NVIDIA-certified engineers will help you architect the optimal configuration for your workload.

Related Chips

More chips in Data Center GPUs.

H100

In Stock

NVIDIA H100 Tensor Core GPU

Enterprise AI training and inference GPU with Transformer Engine and FP8 precision.

ArchitectureHopper

Memory80GB HBM3

TDP700W

H200

In Stock

NVIDIA H200 Tensor Core GPU

Next-gen AI GPU with 141GB HBM3e memory for larger models and faster inference.

ArchitectureHopper

Memory141GB HBM3e

TDP700W

B200

Pre-Order

NVIDIA B200 Tensor Core GPU

Flagship AI GPU with Blackwell architecture, 384GB HBM3e, and 10 TB/s bandwidth.

ArchitectureBlackwell

Memory384GB HBM3e

TDP1000W

GH200

On Order

NVIDIA GH200 Grace Hopper

Superchip combining Grace CPU with Hopper GPU for HPC and AI workloads.

ArchitectureHopper

Memory96GB HBM3 + 480GB LPDDR5X

TDP—