NVIDIA Nemotron 3 Nano 30B MoE model is now available in Amazon SageMaker JumpStart

2026-02-11 · Source: Artificial Intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure · Depth: Intermediate, short

Summary

The NVIDIA Nemotron 3 Nano 30B model, featuring 3 billion active parameters, is now generally available in the Amazon SageMaker JumpStart model catalog. This small language hybrid Mixture-of-Experts (MoE) model is designed for high compute efficiency and accuracy, excelling in coding, scientific reasoning, and math. It leads on benchmarks such as SWE Bench Verified, GPQA Diamond, AIME 2025, Arena Hard v2, LiveCodeBench, BFCL, and IFBench. Nemotron 3 Nano is fully open-source, providing open weights, datasets, and recipes for customization and deployment on various infrastructures. It supports a context window of up to 1 million tokens and functions as a text-based foundation model for both inputs and outputs.

Key takeaway

For AI Engineers building generative AI applications, Nemotron 3 Nano 30B offers a powerful, open-source MoE model with strong performance in coding and reasoning. You should consider deploying it via Amazon SageMaker JumpStart to leverage its managed capabilities and accelerate development without managing complex infrastructure, while also benefiting from its customization options.

Key insights

NVIDIA's Nemotron 3 Nano 30B, an open-source MoE model, offers high efficiency and accuracy for agentic tasks on AWS.

Principles

• Open-source models enable customization and meet privacy needs.
• Hybrid MoE architectures optimize accuracy and compute efficiency.

Method

Deploy Nemotron 3 Nano via Amazon SageMaker JumpStart by selecting the model in SageMaker Studio, deploying to an endpoint, and interacting using AWS CLI or SageMaker SDK.

In practice

• Use for highly-skilled agentic tasks.
• Customize with open weights and datasets.
• Integrate into generative AI applications.

Topics

• NVIDIA Nemotron 3 Nano
• Amazon SageMaker JumpStart
• Mixture-of-Experts
• Generative AI
• Transformer-Mamba Architecture

Code references

• NVIDIA/nim-deploy

Best for: AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• NVIDIA Nemotron 3 Ultra now available on Amazon SageMaker JumpStart — NVIDIA Nemotron 3 Ultra, an MoE model, optimizes cost and speed for complex, long-running AI agent workflows.
• Run NVIDIA Nemotron 3 Nano as a fully managed serverless model on Amazon Bedrock — Nemotron 3 Nano offers an open, efficient, and accurate SLM for specialized agentic AI systems on Amazon Bedrock.
• NVIDIA Nemotron 3 Nano 30B-A3B is Now Available on HexGrid.cloud — Nemotron 3 Nano offers efficient, advanced reasoning and long-context capabilities via a hybrid MoE architecture for practical AI agent deployment.
• NVIDIA Releases Nemotron 3 Super: A 120B Parameter Open-Source Hybrid Mamba-Attention MoE Model Delivering 5x Higher Throughput for Agentic AI — Nemotron 3 Super is an open-source 120B parameter hybrid MoE model for efficient multi-agent reasoning.
• Nvidia Nemotron 3 Nano Omni - First Test and Impression — Nvidia's Nemotron 3 Nano Omni is a fast, multimodal MoE model for local inference and diverse data processing.
• Nemotron 3 Nano: A Very Fast Model That Doesn't Think Too Much — Nemotron 3 Nano is a fast, sparse MoE hybrid model with strong throughput but potential accuracy nuances.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.