Nvidia unveils new AI technologies at annual developer conference

San Francisco: U.S. chipmaker Nvidia unveiled a series of artificial intelligence (AI) innovations spanning graphics, computing infrastructure and sof...

San Francisco: U.S. chipmaker Nvidia unveiled a series of artificial intelligence (AI) innovations spanning graphics, computing infrastructure and software at its four-day annual GTC developer conference, which opened Monday in San Jose, California.

Among them was a new AI graphics rendering technology called DLSS 5, which Nvidia founder and CEO Jensen Huang said can significantly enhance image realism while reducing computational workload.

According to Nvidia, DLSS 5 delivers an AI-powered breakthrough in visual fidelity for games. The system combines traditional 3D graphics data with generative AI models that can predict and automatically fill in missing visual details, enabling GPUs to render richer and more lifelike images.

"Twenty-five years after NVIDIA invented the programmable shader, we are reinventing computer graphics once again," said Huang. "DLSS 5 is the GPT moment for graphics -- blending handcrafted rendering with generative AI to deliver a dramatic leap in visual realism while preserving the control artists need for creative expression."

On the hardware front, Nvidia said it plans to launch its next-generation AI computing system, Vera Rubin, later this year. The company said the system consists of roughly 1.3 million components and is expected to deliver up to 10 times the performance per watt of its predecessor, the Grace Blackwell system.