NVIDIA Unveils New Open-Source AI Tech

As the AI landscape continues to evolve, NVIDIA is furthering its commitment to open-source technologies, unveiling new models for language, robotics, and biology. This move reflects broader industry trends towards democratizing access to AI and fostering innovation. By contributing to the open ecosystem, NVIDIA aims to empower developers worldwide and drive economic growth through efficient reasoning, high-fidelity world generation, and interactive physical AI systems.

The new open models, data, and tools are part of the NVIDIA Nemotron family for AI reasoning, NVIDIA Cosmos platform for physical AI, NVIDIA Isaac GR00T for robotics, and NVIDIA Clara for biomedical AI. These technologies will be made available through Hugging Face, a leading platform for AI model sharing and collaboration. As a top contributor to Hugging Face, NVIDIA has already made over 650 open models and 250 open datasets available, expanding access to cutting-edge AI resources for the global developer community.

“Open models are catalysts to AI innovation, making AI accessible, transparent and responsible,” said Clément Delangue, CEO of Hugging Face. NVIDIA’s contributions to the open model ecosystem will enable millions of developers to build advanced AI applications, driving innovation and growth. Leading software companies, such as ServiceNow, Palantir, and CrowdStrike, are already adopting NVIDIA’s open-source models to power their next-generation AI applications.

The NVIDIA Nemotron family brings ultra-efficient reasoning to specialized AI agents, enabling developers to build intelligent agents for areas like software development, customer service, and IT support. Nemotron Nano 3 and Nemotron Nano 2 VL provide advanced document intelligence, image reasoning, and video analysis capabilities. NVIDIA Cosmos and Isaac GR00T open models and data accelerate the training of robotic systems with humanlike reasoning and cognition.

NVIDIA Clara open models for healthcare and life sciences include Clara CodonFM, which learns the rules of RNA to reveal how changes in its code can improve the design of therapies and medicine. These models will be made available on Hugging Face, build.nvidia.com, and other cloud service providers, enabling developers to build and deploy AI applications with ease.

Source: Official Link