Skip to product information
1 of 1

Cisco

Cisco UCSX-GPU-L40S | UCS X-Series NVIDIA L40S 48GB GPU, 350W, Dual-Slot

SKU:UCSX-GPU-L40S

Request Quote
Sale Sold out
Shipping calculated at checkout.

Stock Status: Enquire

 Trade supply only. Server & GPU quote requests need a company name, ABN and work email address. We do not quote Gmail, Outlook or Hotmail addresses, or supply for home use. 

20-min response

Need a quote or bulk pricing? Answered within 20 minutes, Sydney business hours.

Custom Built Servers

Need a different spec, quantity or configuration?

We build configured-to-order refurbished servers from Cisco, Dell, HPE, NVIDIA Lenovo and Supermicro. Tell us what you need and we'll come back with options within 24 hours.

Configure your server now →

Description

This NVIDIA L40S GPU is a passively cooled data-centre accelerator for the Cisco UCS X-Series modular system. Built on the Ada Lovelace architecture with 48GB of GDDR6 ECC memory, it installs in the UCS X440p PCIe node alongside UCS X210c compute nodes. It suits AI and machine learning inference and training, generative AI and LLM workloads, virtual workstations and VDI, 3D rendering and video processing.

The card is a dual-slot, full-height, full-length design with a PCIe Gen4 x16 interface and 864 GB/s of memory bandwidth. It carries 18,176 CUDA cores, 568 fourth-generation Tensor cores and 142 third-generation RT cores, delivering 91.6 TFLOPS of FP32 performance. Maximum power draw is 350W through a 16-pin connector, and passive cooling relies on chassis airflow.

No licence is needed for basic operation. NVIDIA vGPU and NVIDIA AI Enterprise features require separate NVIDIA licensing. The L40S does not support MIG or NVLink.

Enquire with XS Network Tech for new or used pricing and availability.

Features

- Supported GPU option for the Cisco UCS X440p PCIe node
- Ada Lovelace architecture with 4th-generation Tensor cores for AI inference and training
- 3rd-generation RT cores for ray-traced graphics and rendering
- 48GB of GDDR6 ECC memory for large models and datasets
- Hardware root of trust and secure boot
- vGPU software support for virtual workstations and VDI
- Passive cooling designed for chassis airflow in modular deployments

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

Collapsible content

Technical Specifications

FAQs

Technical Specifications

- GPU: NVIDIA L40S
- Architecture: NVIDIA Ada Lovelace
- GPU memory: 48GB GDDR6 with ECC
- Memory bandwidth: 864 GB/s
- Interface: PCIe Gen4 x16 (64 GB/s bidirectional)
- CUDA cores: 18,176
- RT cores: 142 (3rd generation)
- Tensor cores: 568 (4th generation)
- FP32: 91.6 TFLOPS
- TF32 Tensor Core: 183 TFLOPS (366 TFLOPS with sparsity)
- FP16/BF16 Tensor Core: 362.05 TFLOPS (733 TFLOPS with sparsity)
- FP8 Tensor Core: 733 TFLOPS (1,466 TFLOPS with sparsity)
- Form factor: Dual-slot, full-height, full-length (FHFL)
- Dimensions: 4.4 in (H) x 10.5 in (L)
- Maximum power consumption: 350W
- Power connector: 16-pin
- Cooling: Passive
- Display outputs: 4x DisplayPort 1.4a
- vGPU software support: Yes
- MIG support: No
- NVLink support: No

FAQs

Q: Does this GPU need a licence?
A: No licence is needed for basic operation. NVIDIA vGPU and NVIDIA AI Enterprise features require separate NVIDIA software licensing.

Q: What power does the L40S need?
A: It draws up to 350W through a 16-pin power connector, supplied via the UCS X440p PCIe node. It is passively cooled, so it depends on chassis airflow.

Q: Which Cisco chassis components does it work with?
A: It installs in the Cisco UCS X440p PCIe node, which pairs with UCS X210c compute nodes through X-Fabric. Check the Cisco UCS Hardware Compatibility List for firmware and software version requirements.

Q: Does the L40S support NVLink or MIG?
A: No. The L40S does not support NVLink or Multi-Instance GPU (MIG). It does support NVIDIA vGPU software.

Q: What workloads is this card suited to?
A: It targets AI and machine learning inference and training, generative AI and LLMs, virtual workstations and VDI, 3D graphics and rendering, and video workloads.

  • 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