NVIDIA Ada Lovelace Architecture

Performance and energy efficiency for endless possibilities.

Meet the Next Generation of AI and Graphics GPUs

The NVIDIA Ada Lovelace architecture has been designed to power incredible performance for professional graphics, video, AI, and compute. Defining a dramatically higher baseline of GPU performance, it marks the tipping point for AI, ray tracing, and neural graphics.

Performance—From Data Center to Desk

Data Center GPUs

The architecture’s NVIDIA L4 Tensor Core GPU and NVIDIA L40 GPU accelerate performance for data center workloads.

L4 is optimized for video streaming and inference at scale for a broad range of AI applications, including recommenders, AI assistants, visual search, and contact center automation.

L40 is the most powerful NVIDIA RTX™-enabled GPU for the data center, delivering unprecedented performance for visualization, rendering, virtualization, simulation, and AI.

Workstation GPUs

The NVIDIA RTX 6000 Ada Generation GPU is the ultimate workstation GPU, delivering unprecedented rendering, AI, graphics, data science, and compute performance for professional visualization workloads.

The NVIDIA Ada Lovelace architecture powers a variety of desktops and laptops to maximize productivity and creativity.

Explore the Technologies of Ada

Third-Generation RT Cores

Third-generation RT Cores deliver up to 2X the ray-tracing performance over the previous generation, delivering groundbreaking performance for photorealistic rendering. Enhanced RT Cores combined with Shader Execution Reordering (SER) technology dynamically reorder inefficient workloads, dramatically improving shader performance to accelerate end-to-end ray-traced image rendering performance.

Fourth-Generation Tensor Cores

The Ada Lovelace architecture’s Tensor Cores are designed to accelerate transformative AI technologies like intelligent chatbots, generative AI, natural language processing (NLP), computer vision, and NVIDIA Deep Learning Super Sampling 3.0 (DLSS 3). Ada Lovelace Tensor Cores unleash structured sparsity and 8-bit floating point (FP8) precision for up to 4X higher inference performance over the previous generation. FP8 reduces memory pressure when compared to larger precisions and dramatically accelerates AI throughput.

CUDA Cores

The Ada-based CUDA® cores bring double-speed processing for single-precision floating point (FP32) operations over the previous-generation GPUs. This provides significant performance gains for graphics workflows like 3D model development and compute workflows like desktop simulation for computer-aided engineering (CAE).

Advanced Video and Vision AI Acceleration

NVIDIA Ada Lovelace GPUs take video and vision AI acceleration to the next level and bring our optimized AV1 stack, creating a broad array of new possibilities for use cases like video transcoding, streaming, video conferencing, augmented reality (AR), virtual reality (VR), and vision AI. With up to 2X as many decoders and encoders, combined with the AV1 video format, Ada Lovelace GPUs can host up to 3X more video streams concurrently than the previous generation. On top of this, additional JPEG decoders further speed up applications that need computer vision horsepower.

Deep Learning Super Sampling 3.0 (DLSS 3)

DLSS 3 is a full-stack innovation that delivers a giant leap forward in real-time graphics performance. This breakthrough software leverages the latest hardware innovations within the Ada Lovelace architecture, including fourth-generation Tensor Cores and a new Optical Flow Accelerator (OFA) to boost rendering performance, deliver higher frames per second (FPS), and significantly improve latency. DLSS 3 is a marriage of software and hardware, of AI and dedicated, specialized GPU cores, and is only possible on NVIDIA Ada architecture GPUs.

Data Center Efficiency and Security

The NVIDIA Ada Lovelace L4 and L40 GPUs are optimized for 24/7 enterprise data center operations and designed, built, extensively tested, and supported by NVIDIA and partners for maximum performance, durability, and security. Both the NVIDIA L40 and L4 GPUs feature secure boot with root-of-trust technology, providing an additional layer of security for data centers.

Virtualization-Ready

With next-generation improvements in NVIDIA virtual GPU (vGPU) software, NVIDIA Ada Lovelace data center GPUs can deliver a 1.7x increase in performance for mid- to high-end design workflows running on NVIDIA RTX Virtual Workstation (vWS) and accelerate productivity applications running on NVIDIA Virtual PC (vPC). NVIDIA Ada Lovelace GPUs can support higher user density and resource utilization, and they provide up to 50 percent more GPU memory than the previous generation.