The NVIDIA data center platform is the world’s most adopted accelerated computing solution, deployed by the largest supercomputing centers and enterprises. Whether you’re looking to solve business problems in deep learning and AI, HPC, graphics, or virtualization in the data center or at the edge, NVIDIA GPUs provide the ideal solution. Now, you can realize breakthrough performance with fewer, more powerful servers, while driving faster time to insights and reducing costs.
NVIDIA Data Center GPU
NVIDIA GB200 NVL72
The GB200 NVL72 combines 36 Grace processors and 72 Blackwell GPUs in a rack-mountable enclosure. The GB200 NVL72 is a liquid-cooled solution designed for server racks, featuring an NVLink domain with 72 GPUs that operates as a single, massive graphics processor and delivers 30 times faster real-time LLM inference based on trillions of parameters.
NVIDIA GB200 NVL2
Unmatched performance of a single server.
The NVIDIA GB200 NVL2 platform ushers in a new era of computing for every data center, delivering unmatched performance for mainstream large language model (LLM) inference, vector database searches, and data processing through 2 Blackwell GPUs and 2 Grace processors.
NVIDIA H200
Increasing AI and HPC workloads. The NVIDIA H200 Tensor Core GPU enhances the performance of generative artificial intelligence and high-performance computing (HPC) with revolutionary performance and memory capabilities. As the first GPU with HBM3e, the H200 features larger and faster memory that accelerates generative AI and large language model (LLM) training while also streamlining scientific computations for HPC workloads.
NVIDIA H100
Accelerate any task, anywhere. The NVIDIA H100 is an integral part of the NVIDIA data center platform. Designed for artificial intelligence, HPC, and data analytics, the platform accelerates over 3,000 applications and is available everywhere from the data center to the edge, providing both dramatic performance gains and cost-saving capabilities.
NVIDIA L4
Accelerate workloads related to video, artificial intelligence, and graphics. The NVIDIA L4 Tensor Core GPU, based on the NVIDIA Ada Lovelace architecture, provides universal, energy-efficient acceleration for video, AI, visual computing, graphics, virtualization, and more. Housed in a low-profile design, the L4 is a cost-effective, energy-efficient solution that delivers high throughput and low latency across every server, from the edge to the data center and into the cloud.
NVIDIA L40S
The most powerful universal GPU. Experience breakthrough multi-workload performance with the NVIDIA L40S GPU. Combining powerful AI computing with best-in-class graphics and multimedia acceleration, the L40S GPU is designed to handle next-generation workloads in data centers—from generative artificial intelligence and large language model (LLM) inference and training to 3D graphics, rendering, and video.
NVIDIA A100
Accelerating the most important work of our time. The A100 GPU provides unprecedented acceleration at any scale to power the world’s most efficient flexible data centers for artificial intelligence, data analytics, and HPC. Based on the NVIDIA Ampere architecture, the A100 is the engine of the NVIDIA data center platform.
NVIDIA A2
Versatile entry-level inference. The NVIDIA A2 Tensor Core GPU provides entry-level inference with low power consumption, compact size, and high performance for NVIDIA AI at the edge.
NVIDIA L40
Delivers unprecedented visual computing performance for the data center. From virtual workstation applications to large-scale modeling and simulation—modern visual computing and scientific workflows are growing in both complexity and volume.
NVIDIA A10
The NVIDIA A10 GPU delivers the performance that designers, engineers, artists, and scientists need to meet today’s challenges. This compact, single-slot 150 W graphics processor, combined with NVIDIA Virtual GPU (vGPU) software, can accelerate a wide range of workloads in data centers—from graphics-rich virtual desktop infrastructure (VDI) to artificial intelligence.
NVIDIA A16
Take remote work to the next level with NVIDIA A16. When paired with NVIDIA Virtual PC (vPC) or NVIDIA RTX Virtual Workstation (vWS) software, it enables the creation of virtual desktops and workstations with the power and performance necessary to complete any project from anywhere. Designed specifically for graphics-rich, high-density virtual desktop infrastructure (VDI) and utilizing NVIDIA Ampere architecture.
NVIDIA A30
AI inference and fundamental computing for every enterprise. Enhance the performance of every corporate workload with NVIDIA A30 Tensor Core GPUs. With NVIDIA Ampere architecture Tensor cores and Multi-Instance GPU (MIG) capabilities, it provides secure acceleration for a variety of workloads, including large-scale AI inference and high-performance computing (HPC) applications.
NVIDIA A40
The NVIDIA A40 GPU represents an evolutionary leap in performance and support for multiple data center workloads, combining best-in-class professional graphics with powerful computational acceleration and artificial intelligence to meet today’s design, creative, and scientific challenges. Supporting the next generation of virtual workstations and server workloads, the NVIDIA A40 provides professionals with state-of-the-art features for ray-traced rendering, simulation, virtual production, and more, anytime and anywhere.