Nemotron 3 Nano Omni

Nemotron 3 Nano Omni is an open omni model from NVIDIA that combines reasoning with text, vision, and audio. It is part of the Nemotron line and was released on 2026-04-28. It has a context window of 256,000 tokens and a maximum output of 65,536 tokens. The weights are open. It supports reasoning mode and tool calling. It accepts audio, image, text, and video, and returns text. Pricing starts at $0.10 per million input tokens and $0.25 per million output tokens. It is served by 6 hosts.

Against Nemotron 3 Super: window 262K to 256K, input $0.05 to $0.10.

Range across the hosts that serve it: up to $0.20 per 1M input tokens.

• ·256K token context window
• ·65K token max output
• ·Open weights
• ·Reasoning mode and tool calling
• ·Accepts audio, image, text, video

• Reach for it for multimodal reasoning with text, vision, audio, and video
• Reach for it for long-context tasks up to 256K tokens
• Reach for it for tool-calling agents with open weights

6 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window?

256,000 tokens.

What is the price?

From $0.10 per million input tokens and $0.25 per million output tokens.

What modalities does it accept?

Audio, image, text, and video.