Skip to main content
Loading...

Advantech SKY-TESL-4500BS-32 NVIDIA RTX PRO 4500 Blackwell Server Edition 32GB Single-Slot Passive Server GPU with Multi-Instance GPU

The SKY-TESL-4500BS-32 is an NVIDIA RTX PRO 4500 Blackwell Server Edition server GPU: a single-slot, full-height, headless card with 32 GB of GDDR7 ECC memory, 51 TFLOPS FP32 and up to two MIG instances at 165 W. It suits dense servers running AI video, data processing and streaming where slots and power are limited.

  • Single-Slot Server Edition: A single-slot, full-height, headless card drawing up to 165 W, which suits dense servers where slots and power are limited.
  • Blackwell GPU with FP4 Tensor Cores: 10,496 CUDA cores, 328 fifth-generation Tensor Cores and 82 fourth-generation RT Cores, rated at 51 TFLOPS FP32.
  • 32 GB GDDR7 with ECC: A 256-bit interface gives 800 GB/s of memory bandwidth, with error correction for long-running inference and data processing jobs.
  • Two MIG Instances: Multi-Instance GPU splits the card into up to two instances of 16 GB each, so two services can share one card.
  • Video Engines and PCIe 5.0: Three NVENC and three NVDEC engines for AI video and streaming, on PCIe 5.0 x16 with one PCIe CEM5 16-pin connector.
Form Factor: PCI ExpressPCI Express
USA Support Team: +1 407 214 9446 Call
Select section
  • Overview
  • Specifications
  • Downloads

SKY-TESL-4500BS-32 Overview

Not every server has the slots or power for a 600 W dual-slot card, and many AI video and data processing jobs do not need one. The SKY-TESL-4500BS-32 is the NVIDIA RTX PRO 4500 Blackwell Server Edition from Advantech, an energy-efficient accelerator in a single-slot, full-height card. Advantech positions it for agile data processing, AI video and streaming in data centre, edge and cloud servers.

It carries 10,496 CUDA cores, 328 fifth-generation Tensor Cores with FP4 precision and 82 fourth-generation RT Cores, rated at 51 TFLOPS of FP32 performance. Memory is 32 GB of GDDR7 with ECC on a 256-bit interface, giving 800 GB/s of bandwidth. Multi-Instance GPU can divide the SKY-TESL-4500BS-32 into up to two instances of 16 GB each, which suits running two isolated services on one card. Fourth-generation RT Cores also give it ray-traced rendering where a server handles graphics work as well as AI.

The card connects over PCIe 5.0 x16 and draws up to 165 W through one PCIe CEM5 16-pin connector. It is headless, with no display outputs, and carries three ninth-generation NVENC and three sixth-generation NVDEC engines for encoding and decoding video streams. It relies on the server’s airflow for cooling, and the operating range is 10 °C to 45 °C.

We supply the SKY-TESL-4500BS-32 tested, and fit, configure and burn-in test it in the GPU servers we build, checking airflow and power across every card in the chassis. Where a build needs more memory and compute per card, the SKY-TESL-6000BS-96 is the 96 GB, 600 W Server Edition, and the SKY-QUAD-4500B-32 is the dual-slot workstation version of the RTX PRO 4500 Blackwell with 4x DisplayPort 2.1.

SKY-TESL-4500BS-32 Specifications

Overview
Brand & ModelAdvantech SKY-TESL-4500BS-32
Performance
GPU / AI AcceleratorDiscrete GPU (PCIe)
Supported GPUsRTX PRO 4500 Blackwell Server
AI Performance1600
GPU FamilyNVIDIA RTX PRO Blackwell
GPU ArchitectureNVIDIA Blackwell
PCIe Gen SupportPCIe 5.0
Environmental Conditions
Operating Temperature10°C to 45°C [50°F to 113°F]
Mechanical Properties
Dimensions267mm x 112mm (10.50" x 4.40")

Official Advantech USA Distributor

Things Embedded is an official Advantech USA distributor, and we supply the SKY-TESL-4500BS-32 tested, and fit, configure and burn-in test it in the GPU computers and servers we build. Within the Advantech range, the SKY-TESL-6000BS-96 is the dual-slot 600 W Server Edition with 96 GB of GDDR7 and four MIG instances.

Across our range, the SKY-TESL-4500BS-32 sits with our server GPUs, RTX PRO Blackwell systems and AI edge servers.

Consult. Design. Integrate.

Things Embedded is a leading USA supplier of industrial computing and edge AI systems. We combine technical expertise with a collaborative approach to deliver optimal solutions for your application. Whether it’s an off-the-shelf platform or a custom design, we stay at the forefront of the latest technologies to ensure your success at the edge. Our team works closely with you from concept through deployment, providing guidance, engineering support, and integration expertise every step of the way.

Connect with our solutions experts today to review your requirements and find the ideal solution fit for your application.