• NVIDIA L40 Ada Lovelace GPU with 48GB GDDR6 and 4x DP 1.4
1 1

NVIDIA

NVIDIA 900-2G133-0010-000 | L40 48GB GDDR6 Graphics Card, 300W, 4x DP 1.4

900-2G133-0010-000

Stock Status: Enquire

$14,229.99 inc. GST
$12,936.35 ex. GST
Sale Sold out
Shipping calculated at checkout.

Description

The NVIDIA L40 GPU is a professional data center graphics accelerator built on Ada Lovelace architecture, delivering powerful visual computing, AI training and inference, and ray-tracing capabilities. With 48GB of GDDR6 memory, 18,176 CUDA cores, and third-generation RT cores, the L40 excels in 3D design, rendering, virtual production, video encoding with AV1 support, and deep learning workloads. Designed for 24x7 enterprise data center operations with passive cooling, the dual-slot L40 offers 300W power efficiency and secure boot with root of trust technology for mission-critical deployments.

Features

Ada Lovelace architecture with 5nm process technology for exceptional power efficiency
- 18,176 CUDA cores for parallel processing and general-purpose GPU computing
- Third-generation RT cores delivering up to 2x real-time ray-tracing performance vs previous generation
- Fourth-generation Tensor cores with support for FP8, FP16, BF16, TF32, and INT8 precision for AI acceleration
- 48GB GDDR6 ECC memory for large model support and data integrity in production environments
- 864GB/s memory bandwidth for high-throughput data processing
- Triple NVENC and NVDEC engines with hardware-accelerated AV1, H.265, and H.264 encoding/decoding
- PCIe Gen4 x16 interface with 64GB/s bidirectional bandwidth
- SR-IOV and vGPU support for virtualized multi-tenant deployments
- Four DisplayPort 1.4a outputs for multi-display professional workstations
- Secure boot with root of trust technology for enhanced data center security
- NEBS Level 3 compliance for telecommunications and carrier-grade deployments
- Passive thermal solution optimized for 24x7 enterprise data center operations
- Support for NVIDIA Omniverse, RTX, CUDA, cuDNN, and TensorRT frameworks
- DirectX 12 Ultimate and Vulkan 1.4 support for professional graphics workloads
- Concurrent ray-tracing and shading with hardware-accelerated motion blur

Warranty

All products sold by XS Network Tech include a 12-month warranty on both new and used items. Our in-house technical team thoroughly tests used hardware prior to sale to ensure enterprise-grade reliability.

All technical data should be verified on the manufacturer data sheets.

View full details

specs-tabs

Technical Specifications

FAQs

Technical Specifications

GPU Architecture: NVIDIA Ada Lovelace
- GPU Memory: 48GB GDDR6 with ECC
- Memory Bandwidth: 864GB/s
- CUDA Cores: 18,176
- Tensor Cores: 568 (4th generation)
- RT Cores: 142 (3rd generation)
- FP32 Performance: 90.5 TFLOPS
- RT Core Performance: 209 TFLOPS
- System Interface: PCIe Gen4 x16 (64GB/s bidirectional)
- Display Outputs: 4x DisplayPort 1.4a
- Power Consumption: 300W maximum
- Form Factor: Dual-slot, full-height, full-length (10.5 inches)
- Cooling: Passive thermal solution
- Video Encoding: 3x NVENC with AV1 encode/decode support
- Video Decoding: 3x NVDEC with AV1 support
- Virtualization: SR-IOV and vGPU support
- Security: Secure boot with root of trust
- Compliance: NEBS Level 3 ready
- Dimensions: 267mm (L) x 111mm (H)
- Power Connector: 1x 16-pin PCIe

FAQs

Q: What workloads is the NVIDIA L40 optimized for?
A: The L40 is designed for diverse data center workloads including AI training and inference, ray-tracing rendering, 3D design and simulation, virtual production, video streaming and transcoding with AV1 encoding, NVIDIA Omniverse, cloud gaming, virtual workstations, and professional visualization. Its balanced compute, graphics, and AI capabilities make it suitable for mixed workloads in enterprise data centers.

Q: Does the L40 support virtualization for multi-user environments?
A: Yes, the L40 supports NVIDIA vGPU software for GPU virtualization and SR-IOV for sharing GPU resources across multiple virtual machines. This enables virtual workstation deployments, cloud gaming services, and multi-tenant data center environments where multiple users need GPU acceleration simultaneously.

Q: What are the key differences between the L40 and L40S?
A: The L40S offers higher power consumption (350W vs 300W) and enhanced AI inference performance with Transformer Engine for FP8 precision. Both share 48GB GDDR6 memory, 864GB/s bandwidth, and Ada Lovelace architecture, but the L40S is specifically optimized for large language model inference and generative AI workloads while the L40 balances visual computing and AI capabilities.

Q: Can the L40 handle 8K video encoding and what codecs are supported?
A: Yes, the L40 features triple NVENC and NVDEC engines with hardware-accelerated AV1 encode and decode, supporting up to 8K resolution at 60 FPS in 10-bit color. It also supports H.264 and H.265 codecs, making it ideal for broadcast streaming, video production, transcoding workflows, and real-time video conferencing applications.

Q: What server configurations are compatible with the L40?
A: The L40 is a dual-slot, full-height, full-length PCIe Gen4 x16 card with passive cooling, designed for standard enterprise server chassis with adequate airflow. It is available in NVIDIA-Certified Systems from major OEM vendors and requires a PCIe Gen4 x16 slot, sufficient chassis airflow for passive cooling, and a 16-pin PCIe power connector supporting 300W.

Recently Viewed

  • Request a Quote

    Looking for competitive pricing? Submit a request, and our team will provide a tailored quote that fits your needs.

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Call us Now  

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Contact us