Product Data Sheet / Brochure

NVIDIA RTX A5000
PERFECTLY BALANCED.
BLAZING PERFORMANCE.
Amplified Performance for Professionals
The NVIDIA RTX
A5000 delivers the power, performance, capabilities, and
reliability professionals need to bring their boldest ideas to life. Built on the
NVIDIA Ampere architecture, the RTX A5000 combines 64 second-generation
RT Cores, 256 third-generation Tensor Cores, and 8,192 CUDA
®
cores with 24
GB of graphics memory to supercharge rendering, AI, graphics, and compute
tasks. Connect two RTX A5000s with NVIDIA NVLink
1
to scale memory and
performance with multi-GPU configurations
2
, allowing professionals to work
with memory intensive tasks such as large models, ultra-high resolution
rendering, and complex compute workloads. Support for NVIDIA virtual GPU
software increases the versatility for enterprise deployments.
NVIDIA RTX professional graphics cards are certified with a broad range of
professional applications, tested by leading independent software vendors
(ISVs) and workstation manufacturers, and backed by a global team of support
specialists. Get the peace of mind needed to focus on what matters with the
premier visual computing solution for mission-critical business.
SPECIFICATIONS
GPU memory 24 GB GDDR6
Memory interface 384-bit
Memory bandwidth 768 GB/s
Error-correcting code (ECC) Yes
NVIDIA Ampere architecture-
based CUDA Cores
8,192
NVIDIA third-generation
Tensor Cores
256
NVIDIA second-generation
RT Cores
64
Single-precision performance 27.8 TFLOPS
5
RT Core performance 54.2 TFLOPS
5
Tensor performance 222.2 TFLOPS
6
NVIDIA NVLink Low profile bridges connect two
NVIDIA RTX A5000 GPUs
1
NVIDIA NVLink bandwidth 112.5 GB/s (bidirectional)
System interface PCI Express 4.0 x16
Power consumption Total board power: 230 W
Thermal solution Active
Form factor 4.4” H x 10.5” L,
dual slot, full height
Display connectors 4x DisplayPort 1.4a
7
Max simultaneous displays 4x 4096 x 2160 @ 120 Hz,
4x 5120 x 2880 @ 60 Hz,
2x 7680 x 4320 @ 60 Hz
Power connector 1x 8-pin PCIe
Encode/decode engines 1x encode, 2x decode (+AV1 decode)
VR ready Yes
vGPU software support
7
NVIDIA vPC/vApps, NVIDIA RTX
Virtual Workstation, NVIDIA Virtual
Compute Server
vGPU profiles supported See the Virtual GPU Licensing Guide
Graphics APIs DirectX 12.07
8
, Shader Model 5.17
8
,
OpenGL 4.68
9
, Vulkan 1.2
9
Compute APIs CUDA, DirectCompute, OpenCL
Features
> PCI Express Gen 4
> Four DisplayPort 1.4a connectors
> AV1 decode support
> DisplayPort with audio
> 3D stereo support with stereo
connector
> NVIDIA GPUDirect
®
for Video
support
> NVIDIA virtual GPU (vGPU) software
support
> NVIDIA Quadro
®
Sync II
3
compatibility
> NVIDIA RTX Experience
> NVIDIA RTX Desktop Manager software
> NVIDIA RTX IO support
> HDCP 2.2 support
> NVIDIA Mosaic
4
technology
NVIDIA RTX A5000 | DATASHEET | APR21
To learn more about the NVIDIA RTX A5000, visit www.nvidia.com/rtx-a5000/
1 NVIDIA NVLink sold separately. | 2 Connecting two RTX A5000 cards with NVLink to scale performance and memory
capacity to 48GB is only possible if your application supports NVLink technology. Please contact your application provider to
confirm their support for NVLink. | 3 Quadro Sync II card sold separately. | 4 Windows 10 and Linux. | 5 Peak rates based
on GPU Boost Clock. | 6 Effective teraFLOPS (TFLOPS) using the new sparsity feature. | 7 Display ports are on by default for
RTX A5000. Display ports are not active when using vGPU software. | 8 GPU supports DX 12.0 API, hardware feature level
12 + 1. | 9 Product is based on a published Khronos specification and is expected to pass the Khronos conformance testing
process when available. Current conformance status can be found at www.khronos.org/conformance
© 2021 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, CUDA, GPUDirect, NVLink, Quadro, RTX
Experience, and RTX are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other
countries. Other company and product names may be trademarks of the respective companies with which they are
associated. All other trademarks are property of their respective owners.

Summary of content (1 pages)