NVIDIA L4
GPU architecture NVIDIA Ada Lovelace
Universal GPU accelerator for AI, video, graphics and virtualization
7680 NVIDIA CUDA cores for parallel computing
240 Tensor cores for accelerating AI models and deep learning
60 RT cores for rendering and ray tracing
24 GB of GPU memory with ECC to work with AI models and graphics applications
Memory bandwidth of up to 300 GB/s
Interface PCIe 4.0 x16
Low-profile design and ultra-low power consumption of 72 watts
GPU designed for edge AI, video AI, virtual workstations and cloud
Free shipping from €300
Promocja cenowa na model HDR-15-5
Product intended for professional use only
NVIDIA L4
Description
NVIDIA L4 - energy-efficient GPU accelerator for AI, video and graphics
NVIDIA L4 Tensor Core GPU is a state-of-the-art accelerator designed to support a wide range of workloads in data centres, edge systems and cloud platforms. Based on the NVIDIA Ada Lovelace architecture, the chip delivers high performance for artificial intelligence, video processing, graphics, and virtualisation workloads.
With its compact design, ultra-low power consumption, and high performance, NVIDIA L4 enables GPU infrastructure deployment in a wide range of server systems - from edge computing platforms to large data centres.
NVIDIA Ada Lovelace Architecture
The NVIDIA L4GPU uses the NVIDIA Ada Lovelace architecture to deliver high performance for both AI computing and graphics processing tasks. The accelerator features 7680 CUDA cores, 240 Tensor cores and 60 RT cores to accelerate graphics rendering and deep learning computing.
The card also offers 24 GB of GPU memory with ECC and bandwidth of up to 300 GB/s to support artificial intelligence models, video streams and analytics applications.
CUDA cores
Tensor cores
core RT
GPU memory
GPU for artificial intelligence and video analytics
NVIDIA L4 is widely used in artificial intelligence systems, where high computational performance with very low power consumption is required. The accelerator enables AI models to run in production environments and analyse data in real-time.
video analytics and vision AI
LLM and conversational systems
AI with a knowledge base
VLM and image analysis
transcoding and streaming
AI-as-a-Service platforms
GPU for virtualization and graphics
With support for NVIDIA vGPU technology, the L4 card can be used to build scalable virtual workstation environments and graphics platforms accessible from the cloud.
Accelerator applications include:
- virtual workstations
- rendering 3D graphics
- CAD design platforms
- real-time video processing
- image analysis platforms
Energy-efficient GPU for servers and edge computing
With a low power consumption of just 72 watts and a low-profile design, NVIDIA L4 can be installed in a wide range of server systems and edge computing platforms.
TheGPU is ideal for environments requiring high computing performance with minimal power consumption, such as enterprise AI infrastructures, intelligent video analytics systems and cloud platforms.
Technical Specification
| FP32 | 30.3 teraFLOPs |
|---|---|
| TF32 Tensor Core | 120 teraFLOPS* |
| FP16 Tensor Core | 242 teraFLOPS* |
| BFLOAT16 Tensor Core | 242 teraFLOPS* |
| FP8 Tensor Core | 485 teraFLOPs* |
| INT8 Tensor Core | 485 TOPs* |
| GPU Memory | 24GB |
| GPU Memory Bandwidth | 300 GB/s |
| NVENC | NVDEC | JPEG Decoders | 2 | 4 | 4 |
| Max Thermal Design Power (TDP) | 72 W |
| Form Factor | 1-slot low-profile, PCIe |
| Interconnect | PCIe Gen4 x16, 64 GB/s |
| Server Options | Partner and NVIDIA-Certified Systems with 1–8 GPUs |
| Note | * Specifications subject to change |
