NVIDIA: Nemotron 3 Nano 30B A3B

nvidia/nemotron-3-nano-30b-a3b

164th smartest of 182Cheaper than 95% of paidReasoningToolsJSON

Use via OpenRouter ↗

Intelligence

7.4

164th of 182

Design Elo

—

Speed

—

tokens/sec

Latency

—

first token

Input price

$0.050

15th cheapest

Context

262K

262K max out

How it compares

Smarter than10%

of all ranked models

Cheaper than95%

of all ranked models

Overview

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...

Benchmarks

independent · Artificial Analysis & Design Arena

Artificial Analysis24th percentile

Intelligence Index

7.4

GPQA Diamond

40%

Humanity's Last Exam

SciCode

23%

Tau²-Bench (agentic)

25%

Providers & pricing (3)

Provider	In $/M	Out $/M	Context	Uptime
Crusoefp8	$0.050	$0.200	262K	100%
DeepInfrafp4	$0.050	$0.200	262K	98.6%
Novitafp4	$0.050	$0.200	262K	100%

Specifications

Context window262K

Max output262K

Knowledge cutoff—

Input modalitiestext

Output modalitiestext

Prompt caching—

Cache read price$0.030/M

ModeratedNo

Open weightsnvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 ↗

Nemotron 3 Nano 30B A3B FAQ

How much does Nemotron 3 Nano 30B A3B cost?

Nemotron 3 Nano 30B A3B costs $0.050 per million input tokens and $0.200 per million output tokens via OpenRouter, making it 15th cheapest of 332 paid models.

How smart is Nemotron 3 Nano 30B A3B?

Nemotron 3 Nano 30B A3B scores 7.4 on the Artificial Analysis Intelligence Index, ranking 164th of 182 benchmarked models, with a GPQA Diamond score of 40%.