NVIDIA L40S PCIe GPU 900-2G133-0000-000
Brand: NVIDIA | Category: GPUs
SKU: NVIDIA-900-2G133-0000-000 | Part #: 900-2G133-0000-000 | MPN: 900-2G133-0000-000
Contact for Pricing — Request a Quote
About the NVIDIA L40S PCIe GPU 900-2G133-0000-000
The NVIDIA L40S PCIe GPU 900-2G133-0000-000 from NVIDIA is enterprise-grade GPUs hardware built for data centers. It features 48GB capacity, PCIe interface. The NVIDIA L40S PCIe GPU is a data-center-class accelerator built on the Ada Lovelace architecture, delivering a substantial leap in compute density for both AI workloads and professional visualization. Featuring 48 GB of ECC-protected GDDR6 memory on a 384-bit bus, the L40S combines third-generation RT Cores, fourth-generation Tensor Cores, and NVIDIA CUDA cores in a single PCIe Gen 4 form factor, making it well suited to mixed enterprise environments that must serve AI infe Available in brand new condition. NVIDIA gpus products are trusted by enterprises and data centers worldwide. Available from Omnixon Global. Contact our sales team to request a quote.
Technical Specifications
| Brand | NVIDIA |
| Category | GPUs |
| SKU | NVIDIA-900-2G133-0000-000 |
| Part Number | 900-2G133-0000-000 |
| Condition | New |
| Capacity | 48GB |
| Interface | PCIe |
| Manufacturer Part Number | 900-2G133-0000-000 |
| GPU Architecture | Ada Lovelace |
| CUDA Cores | 18176 |
| GPU Memory | 48 GB GDDR6 ECC |
| Memory Bus Width | 384-bit |
| Memory Bandwidth | 864 GB/s |
| FP32 Performance | 90.5 TFLOPS |
| TF32 Tensor Core Performance | 181 TFLOPS (sparse: 362 TFLOPS) |
| FP16 Tensor Core Performance | 362 TFLOPS (sparse: 724 TFLOPS) |
| FP8 Tensor Core Performance | 1457 TOPS (sparse) |
| RT Core Generation | 3rd Generation |
| Tensor Core Generation | 4th Generation |
| System Interface | PCIe Gen 4 x16 |
| Form Factor | Dual-slot, full-height full-length (FHFL) |
| Thermal Design Power (TDP) | 350 W |
| Cooling | Passive (server chassis airflow required) |
| NVLink Support | Yes — NVLink Bridge (2-way, 96 GB combined) |
| Display Outputs | None (compute and virtual workstation via vGPU software) |
| vGPU Software Support | NVIDIA Virtual PC (vPC), Virtual Applications (vApps), Virtual Workstation (vWS) |
| Hardware Video Encode/Decode | NVENC / NVDEC with AV1 encode support |
| ECC Memory | Yes |
Frequently Asked Questions about NVIDIA L40S PCIe GPU 900-2G133-0000-000
What does the NVIDIA L40S PCIe GPU do?
The NVIDIA L40S PCIe GPU accelerates AI/ML training, inference, scientific HPC, and virtualization (vGPU) workloads. Typical deployments include LLM training clusters, computer-vision pipelines, financial risk modeling, and rendering farms.
What are the headline specs of the NVIDIA L40S PCIe GPU?
Key specifications for the NVIDIA L40S PCIe GPU: capacity 48GB; interface PCIe; 3 years support; new condition; max memory 48GB GDDR6 PCIe GPU optimized for AI inference and professional graphics in mainstream enterprise servers; support 3 years standard (manufacturer); gpu support NVIDIA L40S; ai optimized Yes. Manufacturer part number 900-2G133-0000-000. For the full datasheet with electrical, environmental, and compliance details, contact our pre-sales engineering team.
Is the NVIDIA L40S PCIe GPU compatible with my infrastructure?
The NVIDIA L40S PCIe GPU requires a PCIe Gen4 or Gen5 x16 slot, server power adequate for the card's TDP, and CUDA/ROCm driver support in your hypervisor or bare-metal OS. Sales engineering will confirm chassis fit (1U/2U/4U), PCIe lane count, and PSU headroom before quoting.