NVIDIA L4 Tensor Core GPU

Graphic Cards And Accessories\ Datacenter GPUs

The NVIDIA L4 Tensor Core GPU powered by the NVIDIA Ada Lovelace architecture delivers universal, energy-efficient acceleration for video, AI, visual computing, graphics, virtualization, and more. Packaged in a low-profile form factor, L4 is a cost-effective, energy-efficient solution for high throughput and low latency in every server, from the edge to the data center to the cloud.

Specifications

GPU Features	NVIDIA L4 24GB
GPU Memory	24GB GDDR6
Memory bandwidth	300 GB/s
FP32 Tensor Core	120 teraFLOPS*
BFLOAT16 Tensor Core	242 teraFLOPS*
FP16 Tensor Core	242 teraFLOPS*
FP8 Tensor Core	485 teraFLOPs*
INT4 Tensor Core	485 TOPs*
NVENC \| NVDEC \| JPEG decoders	2 \| 4 \| 4
Max thermal design power (TDP)	72W
Interconnect	PCIe Gen4 x16 64GB/s
Form factor	1-slot low-profile, PCIe

Shown with sparsity. Specifications 1/2 lower
without sparsity.

The breakthrough universal accelerator for efficient video, AI, and graphics.

Fourth-Gen TENSOR
CORES

The Ada Lovelace architecture Tensor Cores are designed to accelerate transformative AI technologies like intelligent chatbots,generative AI, natural language processing (NLP), computer vision, and NVIDIA DLSS
3. Ada Lovelace Tensor Cores unleash structured sparsity and 8-bit floating
point (FP8) precision for up to 4X higher inference performance over the previous generation.

Deep Learning Super
Sampling (DLSS)

NVIDIA DLSS 3 is a revolutionary breakthrough in AI-powered graphics that massively boosts rendering performance. Powered by the new fourth-generation Tensor Cores and NVIDIA Optical Flow Accelerator (OFA) on L4, DLSS 3 uses AI to create additional high-quality frames for graphics-based workloads.

Third-GEN RT
CORES

NVIDIA made real-time ray tracing a reality with the invention of RT Cores, which are processing cores on the GPU specifically designed to tackle performance- intensive ray-tracing rendering. Ada Lovelace’s third-generation RT Cores have twice the ray-triangle intersection throughput, increasing RT-TFLOP performance by over 2X.

Virtualization
Ready

With next-generation improvements in NVIDIA virtual GPU (vGPU) software and 1.5X more GPU memory than the previous generation, L4 increases workstation performance by 1.7X for mid- to high-end design workflows running on NVIDIA RTX™ Virtual Workstation (vWS) and accelerates productivity applications running on NVIDIA Virtual PC (vPC)