Compare AI models

Specifications
Parameters
4B—
Context window
256k128k
Benchmarks
MATH-500
95.4%85.5%
GPQA Diamond
53.2%73.3%
Overview
CompanyNVIDIAOpenAI
Release dateMar 16 2026Sep 12 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Nemotron 3 Nano 4B and o1-preview are evenly matched across the 2 benchmarks they both report (MATH-500, GPQA Diamond). o1-preview shipped 550 days before Nemotron 3 Nano 4B, so benchmark comparisons should account for the intervening progress.

Context windows are 256k (Nemotron 3 Nano 4B) vs 128k (o1-preview). Nemotron 3 Nano 4B is open weight, while o1-preview is closed.

On MATH-500, Nemotron 3 Nano 4B leads at 95.4% vs o1-preview at 85.5%. On GPQA Diamond, o1-preview leads at 73.3% vs Nemotron 3 Nano 4B at 53.2%.