NVIDIA L4 Tensor Core

NVIDIA L4 Tensor Core

NVIDIA L4 Tensor Core GPU: Powering the future of AI in your data center

NVIDIA L4 Tensor Core

NVIDIA's L4 Tensor Core GPU revolutionises data center AI and graphics performance. This next-generation accelerator delivers unparalleled performance for various demanding workloads, from deep learning and machine learning (ML) to graphics processing and remote workstations.

Key Features

  • Excel in AI and graphics workloads - Powered by NVIDIA Ada Lovelace architecture, the NVIDIA L4 delivers exceptional speed for deep learning, ML, video processing, and more - accelerating your workloads and driving faster results.
  • Optimise your data center footprint - The L4's compact design and efficient passive cooling allow for seamless integration into existing server environments, maximising space utilisation and minimising energy consumption.
  • Simplify deployment and management - NVIDIA built the L4 for seamless integration with your data center infrastructure. Its broad software support and low power requirements ensure smooth operation and simplify management, allowing you to focus on core business objectives.

Specifications

Architecture Ada Lovelace
GPU memory 24GB
Memory Bandwidth 300GB/s
Interconnect PCIe Gen4 x16 64GB/s
FP32 30.3 TFLOPS
TF32 Tensor Core 120 TFLOPS with sparsity
BFLOAT16 Tensor Core 242 TFLOPS with sparsity
FP16 Tensor Core 242 TFLOPS with sparsity
FP8 Tensor Core 485 TFLOPS with sparsity
INT8 Tensor Core 485 TOPS with sparsity
Max Thermal Design Power (TDP) 72W
Decoders 2 * NVENC
4 * NVDEC
4 * JPEG
Form Factor PCIe
Single-slot
Low-profile (HH, HL)
Server Options Partner and NVIDIA-Certified Systems with 1-8 GPUs