GPUWorkstationReleased Oct 5, 2020

NVIDIA RTX A6000

48 GB GDDR6 ECC · 384-bit · 10,752 cores

The Ampere generation's workstation flagship, using the fully enabled GA102 with ECC-protected memory and certified drivers for professional rendering, simulation, visual effects and data science. Two cards can be joined over NVLink so that a single application sees a pooled frame buffer, and the card supports NVIDIA's vGPU software for virtualised workstations. Announced in October 2020 and available through workstation partners from that December.

General

Manufacturer: NVIDIA
Model: RTX A6000
Segment: Workstation
GPU chip: GA102
Architecture: Ampere
Release date: Oct 5, 2020
Price: Search on Amazon

Memory

Memory size: 48 GB
Memory type: GDDR6 ECC
Bus width: 384-bit
Bandwidth: 768 GB/s
Memory clock: 2,000 MHz

Clocks & silicon

GPU clock: 1,410 MHz
Boost clock: 1,800 MHz
Shader units: 10,752 CUDA Cores
Tensor / AI cores: 336
RT cores: 84
Transistors: 28.3 billion
Technology (nm): 8 nm
Samsung 8N (8 nm)
TDP / TBP: 300 W

Interface & APIs

Bus interface: PCIe 4.0 x16
SLI / multi-GPU: NVLink (2-way bridge, 112.5 GB/s)
DirectX: 12 Ultimate (12_2)
OpenGL: 4.6
Vulkan: 1.4
Other APIs: CUDA 8.6, OpenCL 3.0, DLSS 2

AI support

Precisions: FP8FP16BF16INT8INT4TF32FP64
AI accelerator: 3rd Gen Tensor Cores (336) + 2nd Gen RT Cores (84)
Peak AI TOPS: —
Notes: 3rd-gen Tensor Cores with sparsity. TF32/BF16/FP16/INT8/INT4 supported; no FP8 on Ampere. FP64 at 1/64 rate.

All NVIDIA GPUs →

Workstation gpus from NVIDIA announced around the same time as the NVIDIA RTX A6000. Add any of them to your compare list to see the specs side by side.

gpu