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
Black and silver PCIe 3.0 x16 card with four GPUs, 32 GB GDDR5, fanless design

NVIDIA

NVIDIA M10 32 GB GDDR5 graphics card

SKU CH-TPQN-H0N4Q MPN M10

A GPU computing processor with four Maxwell GPUs, 32 GB GDDR5 memory, 2650 CUDA cores and fanless design for dense virtualized workloads

  • Memory TypeGDDR5
  • 32 GB GDDR5 memory handles large datasets
  • Four GPUs with 2,650 CUDA cores accelerate parallel workloads
  • PCI Express 3.0 x16 interface provides high bandwidth
  • Fanless design suits dense server deployments
  • ECC memory protects data integrity in compute tasks

Multi-GPU Accelerator for Virtualised Workloads

The NVIDIA Tesla M10 packs four Maxwell GPUs and 32 GB of GDDR5 memory into a single PCI Express 3.0 x16 card. It is built for server environments that need dense GPU compute for virtual desktop infrastructure or remote rendering.

Bandwidth and Power Constraints of PCIe 3.0 x16

The x16 lane width caps theoretical throughput at the PCIe 3.0 ceiling, so sustained compute is limited by that bus rather than the 2650 CUDA cores. A 225 watt draw means the host must supply adequate auxiliary power and cooling without exceeding the slot specification.

Fanless Installation Requires Adjacent Slot

Because the card is fanless it relies on chassis airflow, so an empty adjacent PCIe slot is mandatory for proper thermal dissipation. Installers often forget this clearance and then encounter throttling under sustained load.

Highlights

  • 32 GB GDDR5 memory handles large datasets
  • Four GPUs with 2,650 CUDA cores accelerate parallel workloads
  • PCI Express 3.0 x16 interface provides high bandwidth
  • Fanless design suits dense server deployments
  • ECC memory protects data integrity in compute tasks

Specifications

Memory TypeGDDR5

Questions about this item

Should I pick the 32 GB GDDR5 version or a smaller capacity for virtual desktop workloads?

The M10 ships with 32 GB of GDDR5 split across four Maxwell GPUs, giving each GPU 8 GB for dense virtual desktop or compute tasks.

What does the 2650 CUDA core count deliver in a real server?

The 2650 CUDA cores are distributed over four GPUs, providing hardware-accelerated encoding and parallel compute for multiple simultaneous users.

How does the PCI Express 3.0 x16 interface limit performance?

The x16 Gen 3 link caps host-to-GPU bandwidth at the PCIe 3.0 x16 rate, which can become the bottleneck for heavy frame-buffer transfers.

Also in this aisle

People compared these

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