Nvidia launches open AI models for enterprise agents and PCs

Summarised from 2 outlets · Updated 21 Oct, 12:50 · Archive

Nvidia has launched new open-weight models designed for AI agents in enterprise data centres and on consumer PCs.

Nvidia has launched new open-weight models designed for always-on agents and edge computing. The Nemotron 3.5 Super VL is a large, 120-billion-parameter model intended for agents used by enterprises. Although it uses fewer parameters than Nvidia's Nemotron 3 Ultra model, the company says it can deliver similar accuracy in instruction-following and tool-use tasks.

At the lower end, Nvidia also launched two lightweight Bolt models designed to run locally on new AI PCs running the Nvidia RTX Spark chip, as well as on workstations running the DGX Spark. These consumer-focused devices are being released through partners including Microsoft and Dell, aimed at consumers, programmers and enterprise knowledge workers.

Nvidia claims that the Super VL completes tasks faster than competing models from Anthropic and OpenAI, including Opus 5, GPT 6.1 Sol, Luna and Astra, when post-trained on specific agent tasks such as document intelligence, financial accounting, scientific experimentation and medical insurance claims processing. Nvidia says companies including Skydance, EY, Ramp, Crowdstrike, Perplexity and Salesforce are already using Super VL.

How it is being reported

• Nvidia picks OpenShell to tame AI gentsGive your agents freedom to do useful work by locking them downThe Register · 21 Oct, 12:30
• Nvidia launches new open models to run agents in the enterprise and on PCsNvidia is leaning hard into the trend of enterprise companies wanting to run their own open models in their own data centers and also trying to make on-device, private AI agents a thing.TechCrunch · 19 Oct, 16:37

In this story: Nvidia · Microsoft · Dell · Anthropic · OpenAI · Salesforce

This summary was written by AI from the headlines and standfirsts above, and states as fact only what at least two outlets report. How we use AI · Report a problem