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 M2090 plug-in graphics card, gold edge connector, black metal bracket, dual-slot cooler with radial fan

NVIDIA

NVIDIA M2090 Tesla graphics card 6 GB GDDR5 1.30 GHz 384-bit PCI Express x16

SKU CH-73J8-EK1ZP MPN M2090

Tesla M-class GPU with 6 GB GDDR5, 384-bit bus and 1.30 GHz core for CUDA compute workloads

  • Graphics processorM2090
  • Memory capacity6 GB
  • Interface typePCI Express x16
  • Type of memoryGraphic Card
  • Core frequency1.3 GHz
  • Form factorPlug-in Card
  • 6 GB GDDR5 memory enables large dataset processing
  • 384-bit bus width delivers high memory bandwidth
  • 30 GHz core clock accelerates compute tasks
  • PCI Express x16 interface ensures fast host communication
  • CUDA architecture supports parallel HPC workloads

Tesla M-Class GPU for HPC

The NVIDIA Tesla M2090 is a plug-in card built on the CUDA architecture codenamed Fermi. It targets high performance computing workloads such as seismic processing, biochemistry simulations and computational finance. The board delivers 6 GB of GDDR5 memory across a 384-bit interface with a core clock of 1.30 GHz.

PCI Express x16 bandwidth ceiling

The card connects through a PCI Express x16 slot, so sustained throughput is bounded by that generation’s lane budget. Workloads that stream large datasets across the bus will hit the interface ceiling before the GPU compute units saturate. Plan for host-to-device transfer patterns that keep the pipeline full without exceeding the link capacity.

Power and slot clearance checklist

Verify that the system power supply can deliver the auxiliary connectors this Fermi-class board requires. Confirm the chassis has a full-height, double-slot aperture and enough length for a plug-in card of this class. Ensure the motherboard firmware and operating system expose the device for OpenCL and DirectCompute workloads.

How it compares

Measured against everything we stock in Desktop GPUs & Video Cards, at today’s prices. Recalculated every night.

Price $405.99
Pricier than about 55% of the 4,285 comparable items (median $354.99).
Memory capacity 6 GB
Lower than about 80% of the 3,563 comparable items (median 8 GB).
Core frequency 1.3 GHz
Lower than about 85% of the 1,624 comparable items (median 1.82 GHz).

New to this? How to buy a graphics card with Bitcoin — what to check, what it costs to pay, and what is never asked for.

Specifications

Product
Brand nameNVIDIA
Product lineM
Memory
Type of memoryGraphic Card
Graphics
Graphics processorM2090
Memory capacity6 GB
Core frequency1.3 GHz
Chipset makerNVIDIA
Memory bus width384-Bit
OpenGL supportOpenCL
Ports & slots
Interface typePCI Express x16
Size & build
Form factorPlug-in Card

Questions about this item

Should I choose the 6 GB GDDR5 version of the Tesla M2090 for compute workloads?

The 6 GB GDDR5 frame buffer on the M2090 is sized for large datasets in seismic processing, biochemistry simulations and computational finance where on-board capacity matters.

What does the 1.30 GHz core clock deliver inside a real system?

The 1.30 GHz GPU clock drives the Fermi CUDA cores for double-precision throughput in weather modelling, CAE and CFD tasks that scale with raw compute frequency.

How does the PCI Express x16 384-bit interface affect practical bandwidth?

The 384-bit memory bus paired with GDDR5 feeds the 1.30 GHz core, while the PCI Express x16 host link moves results to the CPU without choking the 6 GB frame buffer.

More in Desktop GPUs & Video Cards

Similar items