All Models

Nvidia Nemotron 3 Nano 30B

nemotronReasoningTool CallingOpen WeightsStructured Output

NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...

Providers3
ReleasedDec 15, 2025
Input Modalitiestext
Output Modalitiestext
Tarsk Usecoding

Benchmarks

6.8
Intelligence Index
13.3
Math Index

Available Providers (3)

ProviderModel IDInput CostOutput CostContextMax OutputDocs
Amazon Bedrocknvidia.nemotron-nano-3-30b$0.06/MTok$0.24/MTok262.1K8.2K
Venice AInvidia-nemotron-3-nano-30b-a3b$0.07/MTok$0.30/MTok128K16.4K
NanoGPTnvidia/nemotron-3-nano-30b-a3b$0.17/MTok$0.68/MTok262.1K235.9K

Capabilities

Reasoning
Tool Calling
Attachments
Open Weights
Structured Output