NVIDIA L40
Unprecedented Visual Computing Performance for the Data Center
From virtual workstations to large-scale modeling and simulation, modern visual computing and scientific applications are growing in complexity and quantity. Enterprises need data center solutions that can deliver extreme performance and scale with versatile capabilities to conquer the diverse computing demands of these increasingly complex workloads.
The NVIDIA L40 delivers unprecedented visual computing performance for the data center, providing next-generation graphics, compute, and AI capabilities to GPU-accelerated applications. Based on the Ada Lovelace GPU architecture, the L40 features third-generation RT Cores that enhance real-time ray tracing capabilities, and fourth-generation Tensor Cores with support for the FP8 data format to deliver over a petaflop of inferencing performance. These new features are combined with the latest generation CUDA Cores and 48GB of graphics memory to accelerate visual computing workload from high-performance virtual workstation instances to large-scale digital twins in NVIDIA Omniverse. With up to twice the performance of the previous generation at the same power, the NVIDIA L40 is uniquely suited to provide the visual computing power required by the modern data center.
Performance Highlights
CUDA Cores
|
18176 |
Tensor Cores
|
568 | Gen 4 |
RT Cores
|
142 | Gen 3 |
GPU Memory
|
48 GB GDDR6 ECC |
Memory Interface
|
384-bit |
Memory Bandwidth
|
864 GB/s |
Display Connectors
|
4x DP 1.4a |
Maximum Digital Resolution
|
4x 5K at 60 Hz | 2x 8K at 60 Hz
4x 4K at 120 Hz | 30-bit Color |
Form Factor
|
4.4" H x 10.5" L | Dual Slot |
Thermal Solution
|
Passive |
Maximum Power Consumption
|
300 W |
Scalable multi-workload performance
Next Generation Graphics
- The NVIDIA L40 brings the highest level of power and performance for visual computing workloads in the data center. Third-generation RT Cores and industry-leading 48 GB of GDDR6 memory deliver up to twice the real-time ray tracing performance of the previous generation to accelerate high-fidelity creative workflows, including real-time, full-fidelity, interactive rendering, 3D design, video streaming, and virtual production — ideal for demanding Media and Entertainment professionals.
Powerful Compute and AI
- Groundbreaking features in the NVIDIA L40 accelerate a wide range of compute-intensive workloads running in the data center, including training, inferencing, data science, and graphics applications. The latest fourth-generation Tensor Cores deliver enhanced AI capabilities to accelerate visual computing workloads and deliver groundbreaking performance for deep learning and inference applications.
Data Center-Ready
- Designed for 24 x 7 enterprise data center operations and optimized to deploy at scale, the NVIDIA L40 utilizes enterprise-grade components, power-efficient hardware, and features like secure boot with internal root of trust. The L40 delivers the highest levels of performance and reliability for data center workloads. Packaged in a dual-slot power-efficient design, the L40 is available in a wide variety of NVIDIA-Certified Systems from leading system builders.
Structural Sparsity
- AI networks are big, having millions to billions of parameters. Not all of these parameters are needed for accurate predictions, and some can be converted to zeros to make the models “sparse" without compromising accuracy. Tensor Cores in the NVIDIA L40 can provide up to 2x higher performance for sparse models. While the sparsity feature more readily benefits AI inference, it can also improve the performance of model training.
Every Deep Learning Framework, 700+ GPU-Accelerated Applications
- The NVIDIA L40 GPU is a key component of the NVIDIA data center platform for deep learning, HPC, and data analytics. It accelerates every major deep learning framework and accelerates over 700 HPC applications. It's available everywhere, from desktops to servers to cloud services, delivering both dramatic performance gains and cost-saving opportunities.
Virtualization Capabilities
- Virtualized compute workloads such as AI, Deep learning, FP32 high-performance computing (HPC), and photorealistic ray-traced graphics, even essential corporate applications and general-purpose productivity software (90% of which utilizes GPU acceleration) can be supported with NVIDIA vApps (vApps), NVIDIA vPC (vPC), and NVIDIA RTX Virtual Workstation (vWS).
Warranty
Free dedicated phone and email technical support
(1-800-230-0130)
Dedicated NVIDIA professional products Field Application Engineers
Resources
Contact pnypro@pny.eu for additional information.