Skip to main content

Technical details of the GPUs

In this article, you will find the technical details and specifications of the GPUs we have installed.

We currently offer the following GPUs for hire:

  • NVIDIA RTX A4000

  • NVIDIA Quadro RTX 6000

  • NVIDIA A100 40 GB

  • NVIDIA A100 80 GB

  • NVIDIA RTX 6000 Ada

NVIDIA RTX A4000

Graphics processor specificationsNVIDIA RTX A4000
GPU memory16 GB GDDR6 with error correction code (ECC)
Display ports4x DisplayPort 1.4
Max. power consumption140 W
Graphics busPCI Express 4.0 x 16
Form factor11.2 cm H x 26.7 cm L, single slot
CoolingActive
Passmark 3D Graphics Mark24,800
Single-precision performance (FP32)19.17 TFLOPS
Number of shading units6,144
Memory interface256-bit
Memory typeGDDR6
Memory bandwidth512 GB/s
ECC supportYes

NVIDIA Quadro RTX 6000

Graphics processor specificationsNVIDIA Quadro RTX 6000
GPU memory24 GB GDDR6
Display ports4x DisplayPort 1.4, 1x USB-C
Max. power consumption260 W
Graphics busPCI Express 3.0 x 16
Form factorFull-size ATX / dual-slot
CoolingActive
3D Graphics Mark score26,300
Half-precision performance (FP32)32.63 TFLOPS
Single-precision performance (FP32)16.30 TFLOPS
Double-precision performance (FP32)0.509 TFLOPS
Number of shading units4,608
Memory bus width384-bit
Memory typeGDDR6
Memory bandwidth672 GB/s
ECC supportYes

NVIDIA A100 40 GB

Graphics processor specificationsNVIDIA A100 40 GB
GPU memory40 GB HBM2e
Display ports/
Max. power consumption250 W
Graphics busPCI Express 4.0 x 16
Form factorFull-size ATX / dual-slot
CoolingPassive (heat sink)
Single-precision performance (FP32)19.5 TFLOPS
Double-precision performance (FP64)9.7 TFLOPS
Number of shading units6912
Memory interface5120 bit
Memory typeHBM2e
Memory bandwidth1555 GB/s
ECC supportYes

NVIDIA A100 80 GB

Graphics processor specificationsNVIDIA A100 80 GB
GPU memory80 GB HBM2e
Display ports/
Max. power consumption300 W
Graphics busPCI Express 4.0 x 16
Form factorFull-size ATX / dual-slot
CoolingPassive (heat sink)
Single-precision performance (FP32)19.5 TFLOPS
Double-precision performance (FP64)9.7 TFLOPS
Number of shading units6,912
Memory interface5,120-bit
Memory typeHBM2e
Memory bandwidth1,555 GB/s
ECC supportYes

NVIDIA RTX 6000 Ada

Graphics processor specificationsNVIDIA RTX 6000 Ada
GPU memory48 GB GDDR6 with error-correcting code (ECC)
Display ports4x DisplayPort 1.4
Max. power consumption300 W
Graphics busPCI Express 4.0 x 16
Form factorFull-size ATX / dual-slot
CoolingActive
Single-precision performance (FP32)91.1 TFLOPS
Number of shading units18,176
Memory interface384-bit
Memory typeGDDR6
Memory bandwidth960 GB/s
ECC supportYes