Nvidia continues to expand its footprint in open source AI on two fronts: acquisitions and new model releases.
The semiconductor giant announced Monday that it has acquired SchedMD, a leading developer of the popular open source workload management system Slurm. Nvidia said it will continue to operate the program, designed for high-performance computing and AI, as open source, vendor-neutral software.
Slurm was originally founded in 2002, and SchedMD was founded in 2010 by lead Slurm developers Morris Jette and Danny Auble. Oubre is the current CEO of SchedMD.
Terms of the deal were not disclosed. Nvidia declined to comment on the news outside of its blog post.
Nvidia has been working with SchedMD for more than a decade and said in a blog post that the technology is a critical infrastructure for generative AI. The company plans to continue investing in technology to “accelerate” access to various systems.
The semiconductor company also released a new family of open AI models on Monday. The company claimed that this group of models, called Nvidia Nemotron 3, is the most “efficient open model family” for building accurate AI agents.
This model family includes the Nemotron 3 Nano, a smaller model for targeted tasks, the Nemotron 3 Super, a model built for multi-AI agent applications, and the Nemotron 3 Ultra, built for more complex tasks.
tech crunch event
san francisco
|
October 13-15, 2026
“Open innovation is the foundation of AI advancement,” Nvidia founder and CEO Jensen Huang said in the company’s press release. “With Nemotron, we are transforming advanced AI into an open platform, giving developers the transparency and efficiency they need to build agent systems at scale.”
In recent months, Nvidia has been ramping up its open source and open AI services.
Last week, the company announced Alpamayo-R1, a new open inference vision language model focused on autonomous driving research. The company also said at the time that it would add workflows and guides covering the Cosmos world model, which is open source under a permissive license, to help developers more effectively use the model to develop physical AI.
This activity reflects Nvidia’s bet that physical AI will be the next frontier for GPUs. Nvidia wants to become the go-to supplier for many robotics (self-driving car) companies looking for AI and software to develop the brains behind the technology.
Source link
