Nvidia persists in broadening its influence in open source AI through two primary avenues: a new acquisition and a model rollout.

The semiconductor powerhouse revealed on Monday that it has acquired SchedMD, the top creator of the widely used open source workload management system Slurm. Nvidia stated that the firm will continue to manage the program, which is tailored for high-performance computing and AI, as open source, vendor-neutral software.

Slurm was first introduced in 2002, with SchedMD founded in 2010 by lead Slurm developers Morris Jette and Danny Auble. Auble currently serves as the CEO of SchedMD.

The specifics of the agreement were not made public. Nvidia chose not to further elaborate on the announcement beyond its blog entry.

Nvidia has collaborated with SchedMD for over a decade and mentioned in its blog that the technology is essential infrastructure for generative AI. The company plans to continue its investments in the technology and “accelerate” its integration with various systems.

Additionally, the semiconductor firm unveiled a new series of open AI models on Monday. The company asserted that this collection of models, named Nvidia Nemotron 3, is the most “efficient family of open models” for creating accurate AI agents.

This series includes the Nemotron 3 Nano, a compact model for specific tasks, the Nemotron 3 Super, designed for multi-AI agent applications, and the Nemotron 3 Ultra, which is aimed at more complex tasks.

“Open innovation serves as the cornerstone of AI advancement,” Jensen Huang, founder and CEO of Nvidia, stated in the company’s press release. “With Nemotron, we’re converting advanced AI into an open platform that provides developers with the transparency and efficiency required to create agentic systems on a large scale.”

Recently, Nvidia has made strides to enhance its open source and open AI initiatives.

Last week, the company introduced a new open reasoning vision language model, Alpamayo-R1, aimed at autonomous driving research. The company also announced that it had included additional workflows and guides related to its Cosmos world models, which are open source under a permissive license, to assist developers in better utilizing the models for physical AI development.

This surge in activity reflects Nvidia’s belief that physical AI represents the next frontier for its GPUs. Nvidia aims to be the preferred supplier for numerous robotics and self-driving vehicle companies seeking the AI and software to develop the underlying intelligence for the technology.