# Nemotron 3 Nano Omni — weights, license and what is actually open

A compact sparse reasoning model that accepts video, audio, images, and text while activating about three billion parameters per token.

Canonical: https://brightaifuture.com/open-models/nemotron
Format: model-family
Source publication: 2026-04-27
Bright publication: 2026-09-19
Substantive update: None recorded
Evidence and review: AI-assisted primary-source check on 2026-09-19; no independent model reproduction or license certification. No independent reproduction of the Nano Omni benchmark suite is attached; performance statements remain vendor-reported.

## Documented release

Nemotron-3-Nano-Omni-30B-A3B-Reasoning

Organization: NVIDIA

NVIDIA technical report publication: 2026-04-27 (day precision). Source: https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Omni-report.pdf

NVIDIA Build availability: 2026-04-28 (day precision). Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Parameters: 33B listed deployment parameters on NVIDIA Build; the same page’s architecture text says 31B A3B, while the model identifier names a 30B-A3B backbone. Accounting differs across first-party fields.

Architecture: Native omni-understanding sparse mixture-of-experts model in the Nemotron 3 family.

Modalities: text, image, audio, video

Context: 262K tokens on NVIDIA Build; the April 27 technical report describes a 256K maximum context

## License and commercial use

NVIDIA Open Model Agreement

Commercial deployment is governed by NVIDIA’s custom Open Model Agreement; users must review its conditions.

## What is open

Weights: available. Official model weights are downloadable from the linked model release. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Architecture: available. The official model page documents parameter activation, modalities, and configuration. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Inference code: available. NVIDIA publishes reference deployment and inference guidance. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Training code: partial. Nemotron and NeMo tooling is public, but complete code for this exact training run was not established. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Training recipe: partial. NVIDIA documents the model’s construction and reasoning behavior without a complete end-to-end reproduction recipe. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Data information: partial. Training approach information is published, but the full multimodal mixture is not established here. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Training data: not-established. No sufficiently specific public artifact was confirmed in this review. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Evaluation: partial. Vendor evaluation results and methods are published on the official model page. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

Commercial use: partial. NVIDIA explicitly permits commercial use under its custom agreement. This is permission with conditions, not a prohibition and not an OSI certification. Source: https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning

## Uses and strengths described in sources

Video and audio understanding

Reasoning

Efficient active parameter count

Long context

## Hardware and quantization

No single minimum is asserted; requirements depend on precision, media length, and runtime.

Check the official model page for supported NVIDIA deployment formats

## Independent evidence

No independent reproduction of the Nano Omni benchmark suite is attached; performance statements remain vendor-reported.

## Limitations

Custom license

Complete multimodal training corpus is not released

Vendor benchmarks need independent reproduction

## Provenance and history

{}

## Original sources

- [Nemotron 3 Nano Omni technical report](https://research.nvidia.com/labs/nemotron/files/NVIDIA-Nemotron-3-Omni-report.pdf)
- [Nemotron 3 Nano Omni 30B-A3B Reasoning](https://build.nvidia.com/nvidia/nemotron-3-nano-omni-30b-a3b-reasoning)
- [Nemotron AI models](https://developer.nvidia.com/topics/ai/nemotron)

## Continue exploring

- [Open Models](https://brightaifuture.com/open-models)
- [Open intelligence thread](https://brightaifuture.com/threads/open)
- [Related evidence record](https://brightaifuture.com/discoveries/nemotron-3-super)
- [Related evidence record](https://brightaifuture.com/discoveries/llama-nemotron-reasoning)