Nvidia launches new open models for enterprise and local agents
Nvidia introduces the 120-billion-parameter Nemotron 3.5 Super VL and lightweight Bolt models to run AI agents locally and in the enterprise.
Stock photo for illustration only, not from the actual event
- Nvidia launches the open-weight Nemotron 3.5 Super VL for enterprises
- Releases two lightweight Bolt models for local AI PCs and workstations
- Claims faster task completion compared to competitors after post-training
- Adopted by major heavy-hitters including Salesforce and EY
Nvidia launched a new suite of open-weight models designed for always-on agents and edge computing, aiming to give organizations more control over their deployments.
The flagship model is the Nemotron 3.5 Super VL, a large 120-billion-parameter model built for enterprise agents. Although it utilizes fewer parameters than the Nemotron 3 Ultra model, Nvidia states it delivers comparable accuracy in instruction-following and tool-use tasks.
Stock photo for illustration only, not from the actual event
At the other end of the spectrum, the company introduced two lightweight Bolt models designed to run locally on new AI PCs powered by the Nvidia RTX Spark chip and workstations running DGX Spark. These computers are being rolled out through partners like Microsoft and Dell.
"Any organization that wants to run agents on Nvidia infrastructure now has an open, optimized model for their tier without having to choose between capability and deployability."
Kari Briski, Nvidia’s vice president of Open Models and Software
Nvidia claims the Super VL completes tasks faster than competing frontier models from Anthropic and OpenAI when post-trained on specific agentic tasks such as document intelligence, financial accounting, and medical insurance claims.
This push into open-weight enterprise models highlights Nvidia's strategy to capture the market where companies prefer local data centers and custom post-training, bypassing proprietary cloud-only labs.
The Bolt lineup includes the Nemotron 3.5 Bolt 74B-A9B 24GB for multistep reasoning and web browsing, alongside the smaller Bolt E4B for conversational agents. Super VL and Bolt 74B are available now, with Bolt E4B launching later this month.
Source: TechCrunch
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment