Skip to content

NVIDIA · Chat

Nemotron 3 Nano Omni

NVIDIA 🇺🇸ChatNemotron familyOpen weights. One of its sources says the weights are publicly available.
From fal.ai's listing:
Open, efficient reasoning model from NVIDIA. 30B A3B hybrid Transformer-Mamba MoE, built for enterprise agentic workflows.
Market median price
$0.4198 blended / 1M
Max context
262K
Listings
5
Providers
5

Evidence and sources

Maker HQ: US (source)
Sources: Crusoe Managed Inference, Fal.ai data, Models.dev dataPrimary data from Crusoe Managed Inference from Fal.ai: primary mode, description from Models.dev: weights availability
Tags from its sources:open-sourcereasoningcodingfunction-callingjson-mode

Advantages and trade-offs

Compared with 10 representative chat models

Advantages

Open weights available

The weights are public. 4 of the 10 peers that say either way have open weights.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 6 maker's prices; 1 market median price.

Input & output

as stated by the model's providers
Text
Images
Video
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured output

Listings

 Eurouternemotron-3-nano-omniSource: EURouter data66K
 CrusoeNemotron 3 Nano Omni · text-imageSource: Crusoe Managed InferenceNot published by the provider$0.30$1.83
 Digitaloceannemotron-3-nano-omniSource: Models.dev data66K$0.50cache read $0.10$0.90
 Nebius AI Studionebius/nvidia/Nemotron-3-Nano-OmniSource: LiteLLM data262K$0.06$0.24

1 listing without a price is hidden.

These providers advertise this model. Researchers have found some resellers serving a different, cheaper model, so see how to check whether a provider meets your needs.

Benchmarks

No benchmark scores yet.

Around the web

  • NVIDIA Nemotron AI Models - NVIDIA Developer
    A55B Paper: Open, Efficient, MoE Hybrid Mamba Transformer Model for Agentic Reasoning Models: Nemotron 3 …A55B Paper: Open, Efficient, MoE Hybrid Mamba Transformer Model for Agentic Reasoning Models: Nemotron 3 Model Collection Datasets: Pretraining and Post training Cookbooks: Deploy Nemotron 3 Ultra on…

    developer.nvidia.com

  • NVIDIA Nemotron 3 Family of Models - NVIDIA Nemotron
    technical report . Nemotron 3 Nano 30B-A3B FP8 : the final post-trained and FP8 quantized …technical report . Nemotron 3 Nano 30B-A3B FP8 : the final post-trained and FP8 quantized Nano model Nemotron 3 Nano 30B-A3B BF16 : the post-trained Nano model Nemotron 3 Nano 30B-A3B Base BF16 : the…

    research.nvidia.com

  • NVIDIA Nemotron 3 Ultra - NVIDIA Nemotron
    NVIDIA Nemotron Projects NVIDIA NVIDIA Follow NVIDIA Nemotron 3 Ultra Published: June 04, 2026 Models …NVIDIA Nemotron Projects NVIDIA NVIDIA Follow NVIDIA Nemotron 3 Ultra Published: June 04, 2026 Models Ultra Tech Report Nemotron 3 Blog Nemotron 3 Ultra Overview We present our most capable model yet…

    research.nvidia.com

  • NVIDIA Nemotron Research - NVIDIA Nemotron
    NVIDIA Nemotron Projects NVIDIA NVIDIA Follow NVIDIA Nemotron Research This website contains technical information about …NVIDIA Nemotron Projects NVIDIA NVIDIA Follow NVIDIA Nemotron Research This website contains technical information about the research that goes into the NVIDIA Nemotron project. The Nemotron project …

    research.nvidia.com

  • Nemotron Training Recipes - NVIDIA Documentation Hub
    nano3 data prep rl --run YOUR-CLUSTER $ uv run nemotron nano3 rl --run YOUR-CLUSTER Note …nano3 data prep rl --run YOUR-CLUSTER $ uv run nemotron nano3 rl --run YOUR-CLUSTER Note : The --run YOUR-CLUSTER flag submits jobs to your configured Slurm cluster via NeMo-Run . See Execution throu…

    docs.nvidia.com

  • Nemotron — NVIDIA NeMo Framework User Guide
    maintained, and its links or information may be outdated or pose security risks. Visit the …maintained, and its links or information may be outdated or pose security risks. Visit the NVIDIA Docs Hub for the latest and most up-to-date content. NVIDIA NeMo Framework User Guide - Home NVIDIA N…

    docs.nvidia.com

  • Llama Nemotron Models — NVIDIA NeMo Microservices Documentation
    variant GPU Count: 8x 80GB Additional Environment Variables: NIM_WORKSPACE : /model-store NIM_PIPELINE_PARALLEL_SIZE : 8 NIM_MAX_MODEL_LEN …variant GPU Count: 8x 80GB Additional Environment Variables: NIM_WORKSPACE : /model-store NIM_PIPELINE_PARALLEL_SIZE : 8 NIM_MAX_MODEL_LEN : 4096 previous Llama Models next Phi Models On this page Ll…

    docs.nvidia.com

  • Llama Nemotron Models — NVIDIA NeMo Microservices Documentation
    quality, performance, cost and safety. Visit the GitHub repo or view the latest docs . …quality, performance, cost and safety. Visit the GitHub repo or view the latest docs . NVIDIA NeMo Microservices Documentation - Home NVIDIA NeMo Microservices Documentation - Home NVIDIA NeMo Micros…

    docs.nvidia.com

  • GitHub - NVIDIA-NeMo/Nemotron: Developer Asset Hub for NVIDIA Nemotron — A one-stop resource for training recipes, usage cookbooks, and full end-to-end reference examples to build with Nemotron models
    Model · Best For · Key Features · Trade-offs · Resources NVIDIA-Nemotron-3-Nano · High-throughput agentic …Model · Best For · Key Features · Trade-offs · Resources NVIDIA-Nemotron-3-Nano · High-throughput agentic workflows, reasoning, tool-use, chat · 31.6B total / 3.6B active (MoE) · Hybrid Mamba-Trans…

    github.com

  • NVIDIA AI Stack — Nemotron
    Skip to main content Back to top Nemotron - Home Nemotron - Home Nemotron Choose …Skip to main content Back to top Nemotron - Home Nemotron - Home Nemotron Choose version Search Ctrl + K GitHub Search Ctrl + K NVIDIA AI Stack NVIDIA AI Stack # Nemotron training recipes are built o…

    docs.nvidia.com

Also known as: nebius/nvidia/nemotron-3-nano-omni, nvidia/nemotron-3-nano-omni

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works