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
Silver passive-cooled GPU card with large finned heatsink and single 8-pin power connector

NVIDIA

NVIDIA Tesla M40 24GB graphics card P8Y46A

SKU CH-JARB-ZCQ6H

Passive-cooled 24 GB GDDR5 compute GPU with 3,072 CUDA cores and 288 GB/s bandwidth for machine-learning workloads

  • Memory InterfacePCI Express 3.0 x16
  • 24 GB GDDR5 memory handles large datasets
  • 3072 CUDA cores accelerate parallel compute workloads
  • 288 GB/s bandwidth feeds data to the GPU fast
  • 250 W passive cooling suits dense server racks
  • PCIe 3.0 x16 interface connects to modern platforms

24GB Tesla M40 for compute

The NVIDIA Tesla M40 is a 24GB GDDR5 graphics card built on the GM200 GPU with 3072 CUDA cores and a 384-bit memory bus. It delivers 7 TFLOPS of single-precision performance and 0.21 TFLOPS of double-precision performance over a PCI Express 3.0 x16 interface. The card draws 250W TDP and uses an 8-pin CPU power connector.

Suits deep learning inference, not display

This accelerator targets machine learning inference and high-performance compute workloads that benefit from 24GB VRAM and 288 GB/s bandwidth. It has no video outputs and uses passive cooling, so it requires server chassis airflow and is unsuitable for workstations needing display connectivity or quiet desktop operation.

Fits dense GPU servers

The dual-slot passive heatsink relies on strong front-to-back system airflow, freeing internal space otherwise occupied by active coolers. The PCIe 3.0 x16 link and single 8-pin power connector simplify cabling in multi-GPU nodes, while the 250W TDP sets a clear thermal budget for rack power planning.

Highlights

  • 24 GB GDDR5 memory handles large datasets
  • 3072 CUDA cores accelerate parallel compute workloads
  • 288 GB/s bandwidth feeds data to the GPU fast
  • 250 W passive cooling suits dense server racks
  • PCIe 3.0 x16 interface connects to modern platforms

Specifications

BrandNvidia
ModelNvidia Tesla M40
Memory InterfacePCI Express 3.0 x16

Questions about this item

Does the passive cooler require a server chassis with forced airflow?

Yes. The card has no fan and relies on high-volume front-to-back airflow from the chassis; typical workstation cases will not provide enough cooling.

What power connector and PSU capacity are needed?

The board draws 250 W and uses a single 8-pin CPU power connector; a quality PSU rated at least 250 W with that connector is required.

Can this card drive a display for local debugging?

No display outputs are present; the GPU is designed for headless compute workloads and cannot connect a monitor directly.

Also in this aisle

People compared these

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