Nvidia is accelerating its expansion of influence in the field of open source artificial intelligence (AI) on two parallel fronts, by acquiring core infrastructure and releasing new-generation open-ended AI models.
This move shows that the semiconductor giant's long-term strategy is to shape the future AI ecosystem, going beyond traditional GPUs.
On Monday (local time), Nvidia announced that it had acquired SchedMD - the company behind Slurm, the most popular open source workload manager system today in the field of high-performance computing (HPC) and AI.
Slurm is widely used in large-scale super computer centers, research laboratories and AI infrastructure worldwide.
According to Nvidia, Slurm will continue to operate as open source software, regardless of the vendor.
The company is committed to maintaining an open development model, while continuing to invest to expand integration with many different AI systems and platforms.
Slurm was launched in 2002, while SchedMD was founded in 2010 by the project's two main developers, Morris Jette and Danny Auble. Auble is currently the CEO of SchedMD.
The financial terms of the deal were not disclosed and Nvidia also declined to comment further in addition to the information posted on the official blog.
Nvidia said it has been collaborating with SchedMD for more than a decade and considers Slurm the foundation infrastructure for generative artificial intelligence.
This acquisition is expected to "accelerate" the process of approaching high-performance AI for many organizations and businesses around the world.
Nvidia also introduced a new open AI model called Nvidia Nemotron 3, described by the company as the most efficient open model line to build accurate AI agents.
Nemotron 3 includes three versions: Nemotron 3 Nano - a compact model for specialized tasks; Nemotron 3 Super, optimized for AI multi-interferent applications; and Nemotron 3 Ultra, designed for complex tasks that require large-scale reasoning and coordination.
open innovation is the foundation for advances in artificial intelligence, emphasized Mr. Jensen Huang, founder and CEO of Nvidia.
According to him, Nemotron helps transform advanced AI into an open platform, bringing transparency and efficiency to developers when building a large-scale AI actor system.
In recent months, Nvidia has continuously strengthened its open AI portfolio.
Last week, the company announced the Alpamayo-R1, an open-ended visual language model, focusing on self-driving vehicle research.
At the same time, Nvidia also added working processes and guidance materials for the Cosmos world model, to support the development of physical AI.
These moves clearly reflect Nvidia's strategic bet on physical AI, a field considered the next step forward for GPUs.
Nvidia is aiming to become a leading AI platform and software provider for autonomous robot and vehicle companies, where smart cars will determine the competitive advantage in the future.
