Click here to LEARN more.

May 28, 2024

NVIDIA L40S GPU: Specs and Use Cases

Paul Painter, Director, Solutions Engineering

What is the NVIDIA L40S?


The NVIDIA L40S GPU is a powerful universal GPU designed for data centers and offers comprehensive acceleration for a wide range of AI-enabled applications. The GPU features the NVIDIA Ada Lovelace Architecture and Fourth-Generation Tensor Cores for efficient model training and inference. 

With hardware support for structural sparsity and optimized TF32 format, it delivers superior performance for AI and data science tasks—while enhancing AI-enhanced graphics capabilities with DLSS for improved resolution in select applications. 

Let’s explore how the L40S accelerates next-generation workloads including generative AI, LLM inference and fine-tuning, rendering, 3D graphics, and video content streaming.


NVIDIA L40S specs


Specification Details
GPU Architecture NVIDIA Ada Lovelace Architecture
GPU Memory 48GB GDDR6 with ECC
Memory Bandwidth 864GB/s
Interconnect Interface PCIe Gen4 x16: 64GB/s bidirectional
NVIDIA Ada Lovelace Architecture-Based CUDA Cores 18,176
NVIDIA Third-Generation RT Cores 142
NVIDIA Fourth-Generation Tensor Cores 568
RT Core Performance TFLOPS 209
FP32 TFLOPS 91.6
TF32 Tensor Core TFLOPS 183 I 366*
BFLOAT16 Tensor Core TFLOPS 362.05 I 733*
FP16 Tensor Core TFLOPS 362.05 I 733*
FP8 Tensor Core TFLOPS 733 I 1,466*
Peak INT8 Tensor TOPS 733 I 1,466*
Peak INT4 Tensor TOPS 733 I 1,466*
Form Factor 4.4″ (H) x 10.5″ (L), dual slot
Display Ports 4x DisplayPort 1.4a
Max Power Consumption 350W
Power Connector 16-pin
Thermal Passive
Virtual GPU (vGPU) Software Support Yes
vGPU Profiles Supported See the virtual GPU licensing guide
NVENC I NVDEC 3x l 3x (includes AV1 encode and decode)
Secure Boot With Root of Trust Yes
NEBS Ready Level 3
MIG Support No
NVIDIA NVLink Support No


What are the NVIDIA L40S features?


Feature Description

Fourth-Generation Tensor Cores


It supports structural sparsity and optimizes the TF32 format for faster AI and data science model training. Accelerates AI-enhanced graphics with DLSS for better performance.


Third-Generation RT Cores


Enhances ray-tracing performance for lifelike designs and real-time animations in product design, architecture, engineering, and construction workflows.


Transformer Engine


Dramatically accelerates AI performance and improves memory utilization for training and inference. Automatically optimizes precision for faster AI performance.


Data Center Ready


Optimized for 24/7 enterprise operations. Designed, built, and supported by NVIDIA. Meets the latest data center standards with secure boot technology for added security.


What industries can benefit the most from the NVIDIA L40S GPU?


The NVIDIA L40S GPU represents a remarkable advancement in computational power. Its exceptional performance and versatility make it a pivotal tool across many industries. 

From healthcare to finance, automotive to robotics, the applications of the NVIDIA L40S are virtually limitless. Here are just a few industries that can benefit substantially from the capabilities of the NVIDIA L40S GPUs.


Industry Use Case



Accelerate your medical imaging tasks such as MRI and CT scans, enabling quicker assessments by radiologists and healthcare professionals.



Excels in computer-aided engineering (CAE) simulations for vehicle design, crash testing, and aerodynamics, leading to optimized designs.


Financial Services


Enhances data analytics for risk analysis, fraud detection, and algorithmic trading, facilitating faster insights and better decision-making.


NVIDIA L40S price


HorizonIQ is committed to delivering exceptional value through our competitive pricing for the NVIDIA L40S GPU. Whether you need a short-term solution or a long-term commitment, our bare metal pricing options can provide you with the GPU you need—without breaking the bank.


NVIDIA L40S 1-Year 2-Year 3-Year

Price per month

$1,250 $1,125 $1,000


HorizonIQ’s NVIDIA L40S on Bare Metal servers


At HorizonIQ, we’re thrilled to offer you the unparalleled power of the NVIDIA L40S GPU directly integrated into our bare metal servers.

With true single-tenant architecture, you get access to the full potential of cutting-edge computational capabilities. This ensures that your workloads receive unmatched acceleration, scalability, and adaptability to tackle the most demanding computing tasks. 

By leveraging our GPU-based bare metal servers, you gain direct access to this state-of-the-art technology, guaranteeing optimal performance, efficiency, and maximum uptime through redundant systems and proactive support. 

With HorizonIQ, you can count on zero lock-in, seamless integration, 24/7 support, and bespoke solutions tailored precisely to your unique requirements.

Ready to accelerate your AI and data center graphic workloads? Contact us today.

Explore HorizonIQ
Bare Metal


About Author

Paul Painter

Director, Solutions Engineering

Read More