We take cryptocurrency — so the coins you hold can buy real hardware.Pay in crypto, spend it on real hardware. Free insured shipping worldwide over $399. Plain, unbranded boxes, sent from Germany.Free insured shipping over $399. Your rate is locked for 30 minutes once checkout opens.Rate locked 30 minutes. Every item sealed, serial-checked and covered by a 24-month warranty.Sealed, 24-month warranty.

Saved items Sign in
NVIDIA Tesla V100S 32GB HBM2 GPU SXM2 module passive cooler

NVIDIA

NVIDIA V100S 32GB HBM2 graphics card

SKU CH-8Y1G-GT82A MPN V100S

A data-centre accelerator with 32 GB HBM2 memory and a passive cooler for AI and HPC workloads

  • Chipset ManufacturerNVIDIA
  • GPUV100S
  • Memory Size32GB
  • Memory TypeHBM2
  • OpenGLOpenCL
  • Form FactorSXM2
  • Accelerate AI workloads with 32 GB HBM2 memory
  • High-bandwidth HBM2 feeds compute-intensive HPC tasks
  • Volta architecture powers mixed-precision training and inference
  • Passive cooler suits dense rack deployments
  • PCI-Express 3.0 x16 provides 16-lane host connectivity

32 GB HBM2 accelerator for AI and HPC

The NVIDIA V100S 32 GB HBM2 card targets data-centre workloads that demand large memory capacity and high bandwidth. It is built for AI training, high-performance computing and graphics acceleration on the Volta architecture. The SXM2 form factor integrates directly into compatible server boards. CUDA, OpenCL, OpenACC and DirectCompute APIs are supported.

32 GB capacity with HBM2 bandwidth

The 32 GB HBM2 memory pool holds large models and datasets without spilling to system memory. High bandwidth keeps compute units fed during dense matrix work and simulation kernels. When models exceed the 32 GB limit, multi-GPU scaling or model parallelism becomes necessary. PCI-Express 3.0 x16 links the card to the host for data transfers.

Passive SXM2 module for dense servers

A passive cooler relies on chassis airflow, eliminating fan noise and moving parts in rack environments. The SXM2 connector provides mechanical retention and high-speed signalling in a compact footprint. This design suits dense GPU nodes where space and power density are constrained. Installation follows the server vendor’s qualified thermal solution.

Highlights

  • Accelerate AI workloads with 32 GB HBM2 memory
  • High-bandwidth HBM2 feeds compute-intensive HPC tasks
  • Volta architecture powers mixed-precision training and inference
  • Passive cooler suits dense rack deployments
  • PCI-Express 3.0 x16 provides 16-lane host connectivity

Specifications

BrandNVIDIA
SeriesTESLA V100S 32GB
ModelTESLA V100S 32GB
Chipset ManufacturerNVIDIA
GPUV100S
Memory Size32GB
Memory TypeHBM2
OpenGLOpenCL
Form FactorSXM2

Questions about this item

Will the V100S 32GB HBM2 card run in a system that only has PCI-Express 3.0 x16 slots?

Yes, the card uses a PCI-Express 3.0 x16 interface, so it will operate in any slot that provides that link width and generation.

How can I verify the V100S 32GB HBM2 GPU is detected and functional after installation?

Check the NVIDIA driver control panel or run nvidia-smi; it should show the V100S GPU with 32 GB of HBM2 memory and list CUDA and OpenCL as supported APIs.

Does the V100S 32GB HBM2 card come with a power cable or mounting bracket, or must I buy them separately?

The card uses a passive cooler and the SXM2 form factor, so it relies on the server chassis for cooling and mounting; cables and brackets are not included and must be sourced for your specific platform.

Also in this aisle

People compared these

Same shelf, same checkout — eight coins and a 30-minute rate lock.