Exxact Features New NVIDIA Ada Generation GPUs - RTX 5000, RTX 4500, RTX 4000, and NVIDIA L40S

Newest NVIDIA Ada Generation GPUs Available in Exxact System Configurators Today

Exxact will feature four new NVIDIA Ada Generation GPUs – the NVIDIA RTX 5000, RTX 4500, RTX 4000, and NVIDIA L40S – announced by NVIDIA at SIGGRAPH 2023, in our workstations, servers, and clusters. The new NVIDIA GPUs deliver compelling gen-over-gen performance for a competitive edge in real-time rendering environments, 3D graphics acceleration and development, high-performance scientific computing, AI and deep learning capabilities, and more. 

Previous generation NVIDIA RTX A5500, A5000, A4500, and A4000 GPUs are still crowd favorites for high-performance computing and the NVIDIA A100 Tensor Core GPU is in high demand to power the artificial intelligence boom. The new Ada Lovelace GPU lineup delivers the performance professionals need to innovate. 

The NVIDIA RTX 5000 and NVIDIA L40S are available to order now in Exxact configurable systems, while the RTX 4500 and RTX 4000 will follow in the fall. Let's delve in to see what to expect from these new GPUs.

The New RTX 5000 Ada Generation vs. RTX A5500

The NVIDIA RTX 5000 improves upon the last generation RTX A5500 with more GPU memory, CUDA cores, RT Cores, and Tensor Cores. And with the newer generation Ada Lovelace architecture, the RTX 5000 is 1.5x faster in AI training, has 2x more throughput in AI Inference, and is 2x more performant in graphics and rendering over the RTX A5500, potentially outperforming the last generation RTX A6000. 

The gen-over-gen performance increases enable every industry to achieve more with RTX 5000. For architectural engineering, product design, and game development workloads, users will be able to build highly complex 3D models, render large environments faster, power immersive high-fidelity VR, and create more with fewer limitations.

RTX 5000 powers research and data science with faster time to completion of complex simulations, quick prototyping and training of dense AI models, deployment of LLMs and generative AI, and execution of scientific discovery through molecular dynamics and computer-aided drug design.

NVIDIA L40S vs. NVIDIA A100

With increased demand in training the next impactful AI model, many startups and corporations gravitate towards NVIDIA DGX and HGX systems as AI complexity increases by the second. The NVIDIA L40S delivers comparable AI performance versus the last generation flagship NVIDIA A100 Tensor Core GPU. The L40S slots in as a mainstream, low barrier of entry, easy-to-implement GPU accelerator for any user to get started and boost their computing power to handle any workload. 

NVIDIA L40S enables 1.2x more generative AI performance and 1.7x more inference performance over the NVIDIA A100. For LLM models, bring 1.2X the performance of an HGX A100 system to the table using an 8-GPU L40S configuration. With performance numbers close to the NVIDIA A100, the L40S’s availability, fast deployment time, and versatility become the deciding factor.

While scalability is a highlight for DGX and HGX systems, the L40S is a competitive option for those on a tighter budget and even tighter timeframe looking to train and deploy AI. And with 48GB of GPU memory, 142 RT Cores, and exceptional single precision performance, the L40S provides versatility as the most powerful universal GPU for the data center, accelerating performance for graphics rendering, video applications, and other various HPC workloads aside from AI. NVIDIA L40S scales best for small and mid-scale operations for training AI, LLM inferencing, and multi-workload environments.

RTX 4500 Ada Generation vs. RTX A4500

The NVIDIA RTX A4500 GPUs are Exxact customer favorites to slot into workstations and servers because of their top-tier performance for the cost. The RTX 4500 Ada Generation, set to release later this year, extends this performance prowess to increase mainstream performance applications like rendering, generative AI, engineering simulations, molecular dynamics, and workloads using the NVIDIA Omniverse platform for 3D creation.

With the increase in memory, cores, and accelerators, RTX 4500 showcases a very healthy performance leap: 1.6x increase in graphics performance, 2.7x increased Omniverse rendering workload, 1.5x in classic rendering workloads, and 1.5x in AI training and inferencing over the last generation RTX A4500.

RTX 4000 Ada Generation vs. RTX A4000

The RTX 4000 will be the single-slot GPU of choice for accelerating graphics, rendering, and high-performance compute workloads. Both RTX 4000 and RTX A4000 showcase the same amount of CUDA cores, RT Cores, and Tensor Cores yet they’re based on district architectures while maintaining similar thermal ratings. Let’s see the gen-over-gen improvements.

The RTX 4000 has a 1.5x performance uplift in graphics and 1.7x rendering performance, and 1.2x generative AI performance, with up to double the ray-tracing throughput and up to triple the Tensor Core throughput. RTX 4000 tackles demanding creative workloads, design, and engineering workflows in a compact single-slot design, perfect for multi-GPU workstation and server configurations.

Where to Get the New NVIDIA Ada Generation GPUs

At Exxact Corporation, we deliver high-performance computing platforms and solutions featuring NVIDIA GPUs; you can expect the RTX 5000 and NVIDIA L40S to show up in the Exxact System Configurator starting today! The RTX 4500 and RTX 4000 will follow shortly after their release.

We strive to provide the computing resources to determine the optimal system for your workload so you can solve complex computing challenges, accelerate groundbreaking discoveries, maintain a competitive edge, and fuel innovation. From a fleet of small form factor workstations all the way to full-scale multi-node clusters, build or upgrade your computing infrastructure with the new NVIDIA Ada Generation GPUs.

Have any questions about the new NVIDIA GPUs? Want to stay up to date on when they become available?
Contact us today for more information!