NVIDIA has released Nemotron 3.5 Lightning, an open-source AI model built for agentic workflows. The move puts the world's largest company by market cap squarely behind open-weight AI, a stance CEO Jensen Huang says is key to America's AI leadership.
What Nemotron 3.5 Lightning Does
The model is designed for agentic workflows — systems where AI doesn't just answer questions but takes actions, like managing tasks, making decisions, or interacting with other software. By releasing it as open-source, NVIDIA is letting developers download, modify, and deploy the model on their own infrastructure, rather than forcing them through a proprietary API.
That's a significant departure from the closed models offered by some rivals. Open-weight models give developers full control over the underlying parameters, which can be fine-tuned for specific industries or use cases. For companies that need to keep data on-premises or want to avoid per-token costs, that flexibility is a big draw.
Huang's Case for Open-Weight AI
Jensen Huang has been vocal about the importance of open-weight models. He said they are key to America's AI leadership, arguing that the ability to inspect, adapt, and build on top of powerful models keeps innovation moving faster than a closed ecosystem ever could.
His comments come as NVIDIA continues to dominate the AI hardware market, with its GPUs powering much of the world's AI training and inference. The company's market cap — now the largest of any company globally — gives its software moves extra weight. When NVIDIA ships an open-source model, it's not just a research project; it's a signal about where the industry is heading.
What the Release Means for Developers
For developers, Nemotron 3.5 Lightning offers a new option in the growing field of open-weight models. It's specifically tuned for agentic tasks, which are becoming more common as businesses look to automate complex workflows. The model can be integrated into existing systems, and because it's open-source, teams can audit its behavior and adapt it to their needs.
That's a different approach from the one taken by some of NVIDIA's partners and competitors, who have kept their most advanced models under lock and key. By going open, NVIDIA is betting that the community will build on its work, improving the model and expanding its use cases in ways NVIDIA alone couldn't achieve.
The release also puts pressure on other AI labs to justify their closed strategies. If the world's most valuable company is giving away its models, why should anyone pay for a black box?
Nemotron 3.5 Lightning is available now. How developers use it — and whether it lives up to its name in real-world agentic tasks — will be the next test.
