A NEW GENERATION OF GRAPHICS PROCESSORS - DEVELOPED FOR THE PROFESSIONAL SECTOR
With the new NVIDIA RTX™ GPUs, artists can create stunning 3D scenes, designers can iterate impressive building architectures in real time, and engineers can develop breakthrough products on any system.
"Hybrid work is the new normal" said Bob Pette, vice president of professional visualisation at NVIDIA. "RTX GPUs, built on NVIDIA's Ampere architecture, deliver the performance for demanding workloads from any device, so people can be productive from anywhere."
For desktops, the new NVIDIA RTX A5000 and NVIDIA RTX A4000 GPUs feature new RT Cores, Tensor Cores and CUDA® Cores to make AI, graphics and real-time rendering up to 2x faster than previous generations.
The new NVIDIA RTX A2000, NVIDIA RTX A3000, RTX A4000 and RTX A5000 laptop GPUs are designed for professionals who require thin and light devices, delivering accelerated performance without compromising mobility. They incorporate the latest generations of Max-Q and RTX technologies and are supported by the NVIDIA Studio™ ecosystem, which includes exclusive driver technologies that enhance creative apps for optimal levels of performance and reliability.
For the data centre, there is the new NVIDIA A10 GPU and A16 GPU. The A10 provides designers and engineers with up to 2.5x the performance of a virtual workstation compared to the previous generation, while the A16 GPU provides up to 2x user density with a lower total cost of ownership and an improved virtual desktop infrastructure experience compared to the previous generation.
Combined with NVIDIA RTX Virtual Workstation (vWS) and NVIDIA Virtual PC (vPC) software, the A10 and A16 GPUs deliver more power, memory and speed to accelerate any workflow, from graphics and AI to VDI, so users can spend more time working efficiently and productively.
NEXT-GENERATION RTX TECHNOLOGY
- Second generation RT cores: Up to 2x the throughput of the previous generation, with the ability to perform ray tracing, shading and denoising tasks simultaneously.
- Third generation Tensor cores: Up to 2 times the throughput of the previous generation, up to 10 times with Sparsity, with support for the new TF32 and BFloat16 data formats.
- CUDA cores: Up to 2.5 times the FP32 throughput of the previous generation for significant increases in graphics and compute workloads.
- Bis zu 24 GB GPU-Speicher: Twice the memory of the previous generation: The RTX A4000 with 16GB of GDDR6 memory and the RTX A5000 with 24GB of GDDR6 memory both support ECC memory for error-free computing. The RTX A5000 is expandable up to 48GB of memory with NVIDIA NVLink® to connect two GPUs together.
- Virtualisierung: Die RTX A5000 unterstützt die NVIDIA RTX vWS Software für mehrere hochleistungsfähige virtuelle Workstation-Instanzen, mit denen Remote-Benutzer Ressourcen gemeinsam nutzen können, um High-End-Design-, KI- und Compute-Workloads zu bewältigen.
- PCIe Gen 4: Doubles the bandwidth of the previous generation and accelerates data transfer for data-intensive tasks such as AI, data science and 3D modelling.
OVERVIEW OF NEW PUBLICATIONS
AVAILABILITY
NVIDIA RTX A4000 GRAPHICS CARD

POWERED BY THE NVIDIA AMP ARCHITECTURE

NVIDIA AMPERE ARCHITECTURE
BASED CUDA CORES
Accelerate graphics workflows with the latest CUDA® cores for up to 2.7x single precision floating point (FP32) performance compared to the previous generation.

SECOND GENERATION
RT COMPUTING UNITS
Create more visually accurate renderings with hardware-accelerated motion blur and up to 2 times faster raytracing performance than the previous generation.

THIRD GENERATION
TENSOR COMPUTING UNITS
Boost training of AI and data science models with up to 11 times faster training performance compared to the previous generation using Sparsity, while integrating advanced features like AI denoising, DLSS and more into your graphics workflows.

16 GB GPU MEMORY
Control graphics- and compute-intensive workflows with 16GB of GDDR6 memory with ECC, twice the memory of the previous generation.

ONE SLOT FORM FACTOR
Take advantage of an energy-efficient, single-slot form factor that fits into a variety of workstation enclosures.

PCI Express Gen 4
Increase data transfer speeds out of CPU memory for data-intensive tasks with support for PCI Express Gen 4.
TECHNICAL DATA
Graphics processor features |
---|
GPU memory Display connections Max. Energy consumption Graphics bus Form factor Cooling VR Ready |
NVIDIA RTX A4000 |
---|
16 GB GDDR6 with error correction code (ECC) 4x DisplayPort 1.4 140 W PCI Express Gen 4 x16 11,2 cm H x 26.7 cm L, one slot Active Yes |
NVIDIA RTX A5000 GRAPHICS CARD

BASED ON THE NVIDIA AMPERE ARCHITECTURE

CUDA CORES BASED ON THE NVIDIA AMPERE ARCHITECTURE
Accelerate graphics workflows with the latest CUDA® cores for up to 2.5x single precision floating point (FP32) performance compared to the previous generation.

SECOND GENERATION RT COMPUTING UNITS
Produce more visually accurate renders faster with up to 2X faster raytracing performance compared to the previous generation and hardware-accelerated motion blur.

THIRD GENERATION TENSOR COMPUTING UNITS
Boost training of AI and data science models with up to 10 times faster training performance compared to the previous generation using Sparsity, while integrating advanced features like AI denoising, DLSS and more into your graphics workflows.

24 GB GPU MEMORY
Tackle memory-intensive workloads, from virtual production to engineering simulation, with 24GB of GDDR6 memory with ECC.Insert Text Here

READY FOR VIRTUALISATION
Turn your personal workstation into multiple high-performance virtual workstations with support for NVIDIA RTX Virtual Workstation (vWS) software.

THIRD GENERATION NVIDIA NVLINK
Scale memory and performance across multiple GPUs with NVIDIA® NVLink™ to handle larger data sets, models and scenes.

PCI Express Gen 4
Increase data transfer speed from CPU memory for data-intensive tasks with support for PCI Express Gen 4.

ENERGY EFFICIENCY
Take advantage of the power-saving dual-slot design, which is 2.5 times more power-efficient than the previous generation and designed for a wide range of workstations.
TECHNICAL DATA
Graphics processor features | NVIDIA RTX A5000 |
---|---|
GPU memory Display ports Max. Power consumption Graphics bus Form factor Cooling NVLink Virtual GPU (vGPU) Software-Support Support for vGPU profiles VR Ready | 24GB GDDR6 with error correcting code (ECC) 4x DisplayPort 1.4* 230 W PCI Express Gen 4 x16 11,2 cm H x 26,7 cm L Dual-Slot Aktive 2-way low-profile (bridges for 2 or 3 slots) NVIDIA Virtual PC (vPC) und Virtual Applications (vApps), NVIDIA RTX vWS, NVIDIA Virtual Compute Server Yes |
SCALABILITY WITH MULTIPLE GRAPHICS PROCESSORS
NVIDIA NVLink Bridge
.jpg)
NVIDIA RTX A6000 GRAPHICS CARDS

CUDA COMPUTING UNITS BASED ON NVIDIA AMPERE ARCHITECTURE
Double speed processing for single precision floating point operations (FP32) and improved power efficiency for significant desktop performance improvements in graphics and simulation workflows, such as complex 3D computer-aided design (CAD) and computer-aided engineering (CAE).
SECOND GENERATION RT COMPUTING UNITS
With twice the throughput of the previous generation and the ability to run ray tracing with either shading or denoising capabilities, the second-generation RT compute units offer tremendous speed advantages for workloads such as photorealistic rendering of movies, architectural design evaluations and virtual prototyping of product designs. This technology also accelerates the rendering of ray-traced motion blur for faster results with greater visual accuracy.
THIRD GENERATION TENSOR COMPUTING UNITS
The new Tensor Float 32 accuracy (TF32) provides 5 times the training throughput of the previous generation to accelerate model training for AI and data science without requiring code changes. Double inference throughput thanks to hardware support for Structural Sparsity. With Tensor compute units, AI also finds application in graphics, with features such as DLSS AI denoising and improved processing in selected applications.
THIRD GENERATION NVIDIA NVLINK®
The increased GPU-to-GPU interconnect bandwidth enables the use of a single, scalable graphics memory to accelerate graphics and computational workloads, as well as handle larger amounts of data.
48 GIGABYTE (GB) GRAPHICS MEMORY
Ultra-fast GDDR6 memory, scalable up to 96GB with NVLink, gives data scientists, engineers and creative professionals the storage volume they need to work with large data sets and workloads such as data science and simulation.
READY FOR VIRTUALISATION
Supported NVIDIA Virtual Graphics Processor(vGPU) software enables the transformation of a personal workstation into multiple high-performance virtual workstation instances, allowing remote users to leverage shared resources for high-value design, AI and compute tasks.
PCI Express Gen 4
PCI Express Gen 4 support doubles bandwidth over PCIe Gen 3 and improves data transfer rates from CPU memory for data-intensive tasks such as AI and data science.
ENERGY EFFICIENCY
With a dual-slot, energy-efficient design, the RTX A6000 is up to twice as efficient as Turing GPUs and is designed to fit into a wide range of workstations from OEM vendors around the world.
TECHNICAL DATA
GPU Features |
---|
Graphics processor memory Display connections Max. Energy consumption Graphics bus Form factor Cooling NVLink vGPU software support vGPU profile support VR Ready |
NVIDIA RTX™ A6000 |
---|
48 GB GDDR6 with error correcting code (ECC) 4x DisplayPort 1.4* 300 W 16x PCI Express Gen 4 11.2 cm (H) x 26.7 cm (L); double slot Aktiv 2-way low-profile (bridges for 2 or 3 slots) for networking 2 RTX A6000s NVIDIA vPC/vApps, NVIDIA RTX Virtual Workstation, NVIDIA Virtual Compute Server Yes |
SCALABILITY WITH MULTIPLE GRAPHICS PROCESSORS
NVIDIA NVLINK BRIDGE
.jpg)
NVIDIA A16 GRAPHICS PROCESSOR
Let us notify you as soon as the A16 is available. Please use our notification form.

FEATURES
DEVELOPED FOR ACCELERATED VDI
Offers a quad-GPU board design that is optimised for user density and, when combined with NVIDIA vPC software, enables graphics-intensive virtual PCs that can be accessed from anywhere
SUPERIOR USER EXPERIENCE
Provides a higher frame rate and lower end-user latency than a CPU-only VDI. This provides more responsive applications and a user experience indistinguishable from that of a native PC.
DOUBLE USER DENSITY
Designed specifically for graphics-intensive VDI, with support for up to 64 concurrent users per board in a dual-slot form factor
HIGH RESOLUTION DISPLAY
Supports multiple high-resolution monitors (up to two 4K or a single 5K monitor) for maximum productivity and photorealistic quality in a virtualisation environment
MORE THAN DOUBLE ENCODER THROUGHPUT
More than twice the encoder throughput of the previous generation M10 provides the performance for video and multimedia streaming for multiple users
VIDEO IN HIGHEST QUALITY
Supports the latest codecs for the highest quality video experiences, including H.265 encoding/decoding, VP9 and AV1 decoding
PCI Express Gen 4
Supports transfer from CPU memory at PCI Express Gen 4 speeds for data-intensive tasks
NVIDIA AMPERE ARCHITECTURE
Leverages NVIDIA Ampere architecture-based CUDA® compute units, second-generation RT compute units and third-generation Tensor compute units. These provide the flexibility to host virtual workstations powered by NVIDIA RTX™ Virtual Workstation software (vWS) or use unused VDI resources to run compute workloads with NVIDIA Virtual Compute Server (vCS).
GRAPHIC PROCESSOR SPECIFICATIONS
Features | A16 |
---|---|
Graphics processor memory | 4x 16 GB GDDR6 with error correcting code (ECC) |
GPU memory bandwidth | 4x 232 GB/s |
Max. Energy consumption | 250 W |
Connectivity | 16x PCI Express Gen 4 |
Form factor | FHFL-Dual-Slot (Full Height, Full Length) |
Cooling | Passiv |
vGPU software support | NVIDIA Virtual PC (vPC), NVIDIA Virtual Applications (vApps), NVIDIA RTX Workstation (vWS), NVIDIA Virtual Compute Server (vCS) |
vGPU profile support | View virtual GPU licensing guide |
NVENC | NVDEC | 4x | 8x (with AV1 decoding) |
Secure and controlled booting with HW root-of-trust | Yes |
Ready for NEBS | Level 3 |
Power connection | 8-pin CPU |
EGX PLATFORM FOR PROFESSIONAL VISUALISATION

NVIDIA A10 TENSOR CORE GPU
PROFESSIONAL FEATURES

REMOTE COLLABORATION
Take your virtual workplaces to the next level and enable a new era of collaboration with NVIDIA OmniverseTM, an open virtual collaboration platform. Transform complex visual workflows in creativity, design and engineering by connecting users and teams to design tools, assets and projects for collaborative iterations in a virtual world.

ENTERPRISE-CLASS
Virtual workstations with NVIDIA RTX vWS software offer the same optimal performance and stability as physical workstations. Because they use RTX Enterprise drivers, virtual workstations benefit from extensive testing across a wide range of industry applications and certifications from over 100 independent software vendors (ISVs).

3D VISUALISATION
Run professional 3D visualisation applications in the data centre with NVIDIA RTX™ Virtual Workstation (vWS) software - the only virtual workstation that supports NVIDIA RTX technology. RTX enables incredible 3D designs, photo-realistic simulations and stunning visual effects faster than ever before.

KI ACCELERATED APPS
Bring the power of AI to visual computing with NVIDIA RTX. RTX enables developers to create AI-enhanced applications that deliver unprecedented workflow acceleration to end users. This dramatically accelerates the creativity of artists and designers by freeing up time and resources through intelligent manipulation of images, automation of repetitive tasks and optimisation of computationally intensive processes.

INTERACTIVE RENDERING
Achieve fully interactive, photorealistic visualisation in the application viewport by connecting to one or more servers to boost virtual desktop performance. By using a shared pool of GPU-accelerated data centre resources to run virtual workstations and render nodes, productivity is increased and projects can be completed faster.
A10 TECHNICAL DATA AND FEATURES
Features |
---|
FP32 TF32 Tensor Core BFLOAT16 Tensor Core FP16 Tensor Core INT8 Tensor Core INT4 Tensor Core RT Core Encode/decode GPU memory GPU memory bandwidth Interconnect Form factors Max thermal design power (TDP) vGPU software support *With Sparsity |
A10 |
---|
31.2 teraFLOPS 62.5 teraFLOPS | 125 teraFLOPS* 125 teraFLOPS | 250 teraFLOPS* 125 teraFLOPS | 250 teraFLOPS* 250 TOPS | 500 TOPS* 500 TOPS | 1,000 TOPS* 72 RT Cores 1 encoder 2 decoder (+AV1 decode) 24GB GDDR6 600GB/s PCIe Gen4 64GB/s Single-slot, full-height, full-length (FHFL) 150W NVIDIA Virtual PC, NVIDIA Virtual Applications, NVIDIA RTX Virtual Workstation, NVIDIA Virtual Compute Server |
EGX PLATFORM FOR PROFESSIONAL VISUALISATION
.jpg)