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
Full-length dual-slot card with passive heatsink and 12 GB GDDR5 memory

NVIDIA

NVIDIA Tesla K40 12 GB GDDR5 Passive Cooler video card accessory

SKU CH-MYXS-GAE13

Computational accelerator with 2880 CUDA cores, 12 GB GDDR5 memory and passive cooling for server workloads

  • SLI Bridge3-Way
  • Option1 Slot Spacing HB SLI
  • SpecificationsHP 747401-001 F1R08A NVIDIA Tesla K40 12GB Computational Acc
  • Features2880 CUDA cores | 1.43 Tflops of double-precision peak perfo
  • 2880 CUDA cores enable massive parallel computing
  • 12 GB GDDR5 memory handles datasets twice as large
  • 43 Tflops double-precision performance accelerates scientific models
  • 384-bit bus width feeds data at 3000 MHz
  • Passive cooler suits dense server configurations

Tesla K40 12 GB GDDR5 accelerator

The NVIDIA Tesla K40 is a computational accelerator with 12 GB of GDDR5 memory and 2880 CUDA cores. It delivers 1.43 Tflops double-precision and 4.29 Tflops single-precision peak performance for scientific modelling and big data analytics. The card uses a passive cooler and requires a full-length, full-height dual-slot space in the host server. It is built for GPU Boost to convert power headroom into user-controlled performance.

Passive cooling and server airflow dependency

The passive cooler has no fan, so the card relies entirely on directed server airflow for thermal management. A 384-bit memory bus runs at 3000 MHz while the core operates at 745 MHz, but sustained clocks depend on case pressure and inlet temperature. Without adequate chassis airflow the GPU will throttle, reducing the 10x CPU speed-up claimed for GPU Boost workloads. Buyers without controlled server environments cannot meet the cooling need.

Full-length dual-slot installation constraints

The full-length, full-height form factor demands a compatible server chassis with a dual-slot opening and strong front-to-back airflow. The 3-way SLI bridge with 1-slot spacing is included, but multi-GPU scaling requires matching board spacing and power distribution not guaranteed in all racks. Installers often find that cable routing or adjacent cards block the required airflow path. Workstations lacking server-grade cooling cannot accommodate this card.

Highlights

  • 2880 CUDA cores enable massive parallel computing
  • 12 GB GDDR5 memory handles datasets twice as large
  • 43 Tflops double-precision performance accelerates scientific models
  • 384-bit bus width feeds data at 3000 MHz
  • Passive cooler suits dense server configurations

Specifications

BrandHP
ModelF1R08A
Part Number747401-001
SLI Bridge3-Way
Option1 Slot Spacing HB SLI
SpecificationsHP 747401-001 F1R08A NVIDIA Tesla K40 12GB Computational Accelerator
Features2880 CUDA cores | 1.43 Tflops of double-precision peak performance | 4.29 Tflops of single-precision peak performance

New to this? Which coin should you pay with — what to check, what it costs to pay, and what is never asked for.

Also in this aisle

People compared these

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