Newsletter Subscribe
Enter your email address below and subscribe to our newsletter
[forminator_form id="25163"]

cnbc+1blogs.nvidiacnbcNvidia on Tuesday released Nemotron 3.5 Lightning, a free, open-source AI model built for enterprise use that can run on a single GPU, alongside a new model routing tool called NeMo Switchyard designed to cut costs on AI workloads.
The release marks Nvidia's first open-source model since CEO Jensen Huang posted an open letter on X in late July defending open AI models and urging the U.S. government to avoid restrictions that could push innovation overseas.cnbc
Nemotron 3.5 Lightning is a 30-billion-parameter mixture-of-experts model designed for high-volume, specialized tasks within larger multi-agent systems. Nvidia said the model delivers up to four times faster output speed and 30% faster agentic task completion compared with other models in its class.blogs.nvidia+1
The model is fully customizable and can be post-trained on an organization's own data. Kari Briski, Nvidia's vice president of generative AI software, told the Wall Street Journal News Corp that the model is "remarkably easy to customize." She cited CodeRabbit as an example: the company used Nvidia's standard training recipe and produced a router agent for $85 in about two hours. Companies including CrowdStrike and Harvey have also tested and customized the model.siliconangle+1
For Nvidia, open-source AI serves a clear business interest. "Free AI should be great for hardware," Huang told Axios last month. "Free AI should be great for chips."cnbc
Alongside Lightning, Nvidia released NeMo Switchyard, an open-source model routing library that automatically directs each AI agent request to the most suitable model based on quality, latency and cost. Internal benchmarks showed Switchyard maintained frontier-level accuracy while reducing task completion cost to nearly one-third of using a single frontier model alone.blogs.nvidia
Partners reported early results: Ramp matched frontier model performance while cutting costs by 58% and runtime by 33%, while LangChain achieved 74% lower cost by routing only 7% of calls to a frontier model.developer.nvidia+1
The release comes one day after Meta Platforms CEO Mark Zuckerberg published a manifesto arguing for open-source AI, as Meta released a new coding model. Nvidia's move also follows Huang's entry into a policy debate sparked by China's Kimi K3 model, which raised concerns in Washington about intellectual property theft through distillation techniques.cnbc
Nemotron 3.5 Lightning is available on Hugging Face, ModelScope, OpenRouter and Nvidia's own platforms. Nvidia is also already developing a next-generation Nemotron 4 model, according to The Information.tradingview+1