Compare AI models

Specifications
Parameters
—4B
Context window
128k128k
API pricing
Cheapest input
$0.44Mistral—
Cheapest output
$2.20Mistral—
$198Mistral—
Benchmarks
IFEval
89.4%75.5%
MATH-500
91%96.2%
GPQA Diamond
57.1%55.1%
Overview
CompanyMistralNVIDIA
Release dateMay 7 2025May 20 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Medium 3 leads Llama Nemotron Nano 4B v1.1 on 2 of the 3 benchmarks they both report (IFEval, MATH-500, GPQA Diamond). Mistral Medium 3 shipped 13 days before Llama Nemotron Nano 4B v1.1, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Mistral Medium 3) vs 128k (Llama Nemotron Nano 4B v1.1). Mistral Medium 3 is closed, while Llama Nemotron Nano 4B v1.1 is open weight.

On IFEval, Mistral Medium 3 leads at 89.4% vs Llama Nemotron Nano 4B v1.1 at 75.5%. On MATH-500, Llama Nemotron Nano 4B v1.1 leads at 96.2% vs Mistral Medium 3 at 91%. On GPQA Diamond, Mistral Medium 3 leads at 57.1% vs Llama Nemotron Nano 4B v1.1 at 55.1%.