We’re excited to announce that NVIDIA Nemotron™ 3 Nano Omni, an extremely efficient multimodal model with breakthrough reasoning and understanding, is now live on Vultr. It can be deployed on dedicated GPU clusters or through Vultr Serverless Inference, accelerated by NVIDIA Dynamo 1.0 inference software.

NVIDIA Nemotron™ 3 Nano Omni accelerates agentic AI tasks with industry-leading accuracy. It powers sub-agents with advanced reasoning across audio, video, images, documents, and text in enterprise agent systems – all within one simple reasoning loop.

NVIDIA Nemotron™ 3 Nano Omni delivers higher throughput and less overhead management at a lower cost. The open model is fully customizable, transparent, and controllable.

It’s the latest chapter in a long partnership between Vultr and NVIDIA, two companies that believe openness drives the most impactful adoption and innovation. NVIDIA’s open ecosystem and Vultr’s composable cloud infrastructure together ensure that developers can unlock new levels of agentic performance and efficiency without unnecessary lock-in.

Vultr’s global cloud infrastructure, with 33 data center regions and a robust security and compliance portfolio, enables worldwide access to cutting-edge GPU acceleration. And with Vultr’s NVIDIA Exemplar Cloud validation, builders everywhere can feel confident in the reliability and consistency of their workload performance.

Get started

Want to start building with NVIDIA Nemotron™ 3 Nano Omni? Get started today with the "nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16" model name.