NVIDIA NASDAQ: NVDA used its 2026 SIGGRAPH Research Keynote to outline a broad push to combine computer graphics, simulation and artificial intelligence, including a new DLSS 5 technology for real-time rendering, advances in AI-assisted physics simulation and new additions to its Cosmos world foundation model platform for physical AI.
The keynote opened with NVIDIA framing computer graphics as entering “a new era,” with AI increasingly tied to rendering, simulation, robotics and digital twins. Jensen, who introduced the session, said NVIDIA’s history at SIGGRAPH has included programmable GPUs, CUDA, RTX and Omniverse, and argued that virtual worlds will be central to training robots before they operate in the real world.
“Before robots operate in the real world, they will learn in virtual worlds with synthetic experiences,” Jensen said. “That is why computer graphics matter more than ever.”
DLSS 5 Targets Real-Time Photorealism
Edward Liu, NVIDIA’s Director of Applied Deep Learning Research and the technical leader behind DLSS, introduced DLSS 5, describing it as a new generation of the company’s AI rendering technology. Liu said DLSS 5 uses traditional rendering as a foundation, then applies generation to enrich the final appearance of the image in real time.
“The renderer keeps building the world exactly as the game has authored it,” Liu said. “The generation becomes the learned stage afterwards to enrich its appearance.”
Liu said DLSS 5 is intended to combine the controllability of rendering with the photorealistic knowledge learned by generative models. He emphasized that the technology is not designed to replace graphics pipelines, but to extend them. He described DLSS 5 as adding a third category of AI use in real-time rendering, alongside reconstruction and function approximation.
According to Liu, NVIDIA had to address three core challenges: preserving artistic intent, maintaining temporal coherence frame by frame and fitting within the tight performance budget of real-time games. He said the model uses renderer outputs and internal buffers such as albedo, surface normals and lighting information to preserve details that are important to a scene, while enhancing elements such as subsurface scattering, material response, contact shadows and environment lighting.
Liu said DLSS 5 runs causally, “one frame in, one frame out,” without looking ahead, and was distilled into a smaller one-step pixel-space diffusion transformer model focused specifically on making real-time rendering appear more realistic. He said DLSS 5 is “shipping this fall.”
Artists Get Controls Over AI-Enhanced Frames
Gaff, described as a creative artist, demonstrated how developers and artists can direct DLSS 5. He said the technology respects the original rendered frame and does not change geometry, but can uplift images by improving contrast, ambient occlusion, contact shadows, reflections and subsurface scattering.
Gaff showed controls including different models, structure intensity and tone intensity. He said developers can choose different models for different scenes or cut scenes, and can use masks to apply DLSS 5 effects to specific characters, props or parts of an environment.
“DLSS 5 is fully controllable from the developer,” Gaff said, adding that NVIDIA is working with partners to incorporate feedback so the technology can serve artists, art directors and creative directors.
NVIDIA Highlights AI Physics for Simulation
Neil Ashton discussed physics-based simulation and how AI could help reduce the computational cost of high-fidelity simulations. He pointed to a large climate simulation running on more than 20,000 GPUs at one-kilometer resolution and a 50 billion-cell grid, calling it an example of the accuracy possible with physics-based methods but also a reminder of their cost.
Ashton said AI models trained on simulation data are already being used in weather and climate, where they can predict future weather in seconds or minutes compared with hours or days. He said weather centers now use AI models in production, and highlighted StormScope as an advanced AI model trained on satellite and observation data for storm prediction.
He also described applying similar methods to engineering simulations, such as airflow over aircraft. Ashton said an open dataset of roughly 2,000 aircraft simulations generated about 200 terabytes of data, while the trained model checkpoint was about 200 megabytes. He said the model could predict unseen geometries or boundary conditions more than 10,000 times faster, with accuracy within about 1% or 2%.
Cosmos Platform Expands for Physical AI
Ming Liu, VP of the Cosmos Lab at NVIDIA, said physical AI faces a data problem because robots need to learn from the real world, but real-world data is slow to collect. He described Cosmos as NVIDIA’s world foundation model for physical AI developers, designed to provide better data, better environments and better starting points.
Liu said Cosmos can support world understanding, prediction, simulation and action using one shared representation, based on the idea that physical AI tasks draw from the same physics. He described a mixture-of-transformers architecture with an autoregressive tower for reasoning and a diffusion tower for generation, aligning language, vision, audio and action.
Liu announced Cosmos 3 Edge, a four-billion-parameter model built to run real time on devices such as Jetson Thor, RTX and DGX Spark. He said it is intended to enable robot policy and video analytics without a round trip to a data center. NVIDIA also demonstrated a robot arm and camera connected to Jetson Thor running Cosmos 3 Edge policy for real-time control.
Liu also announced Cosmos Dreams, described as neural closed-loop simulators. The first version is designed for autonomous vehicles, generating what vehicle sensors will see based on actions taken by a policy model. In a live demo, Andy showed an autonomous driving simulation generated from a single frame and controlled with a PS5 controller, running on a single RTX 6000 Ada Generation workstation GPU.
Liu said Cosmos Dreams can be used for policy verification and training by generating scenarios that are difficult to craft in the real world. He said Cosmos is being used across NVIDIA efforts including Metropolis VSS, Isaac, Optane and GR00T, and invited developers and partners to join the Cosmos platform.
About NVIDIA (NASDAQ:NVDA)
NVIDIA Corporation, founded in 1993 and headquartered in Santa Clara, California, is a global technology company that designs and develops graphics processing units (GPUs) and system-on-chip (SoC) technologies. Co-founded by Jensen Huang, who serves as president and chief executive officer, along with Chris Malachowsky and Curtis Priem, NVIDIA has grown from a graphics-focused chipmaker into a broad provider of accelerated computing hardware and software for multiple industries.
The company's product portfolio spans discrete GPUs for gaming and professional visualization (marketed under the GeForce and NVIDIA RTX lines), high-performance data center accelerators used for AI training and inference (including widely adopted platforms such as the A100 and H100 series), and Tegra SoCs for automotive and edge applications.
This instant news alert was generated by narrative science technology and financial data from MarketBeat in order to provide readers with the fastest reporting and unbiased coverage. Please send any questions or comments about this story to contact@marketbeat.com.
Before you consider NVIDIA, you'll want to hear this.
MarketBeat keeps track of Wall Street's top-rated and best performing research analysts and the stocks they recommend to their clients on a daily basis. MarketBeat has identified the five stocks that top analysts are quietly whispering to their clients to buy now before the broader market catches on... and NVIDIA wasn't on the list.
While NVIDIA currently has a Moderate Buy rating among analysts, top-rated analysts believe these five stocks are better buys.
View The Five Stocks Here
Looking to profit from the electric vehicle mega-trend? Click the link to see our list of which EV stocks show the most long-term potential.
Get This Free Report
