Compare Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) with measured models

A reliable intelligence-versus-cost verdict requires both models' intelligence scores and benchmark-run costs. Missing data is not a performance verdict.

Measured matchups