Compare AI models

Specifications
Parameters
—550B
Context window
128k1M
API pricing
Cheapest input
—$0.40DeepInfra
Cheapest output
—$1.76DeepInfra
—$210KT
Benchmarks
GPQA Diamond
65.7%86.7%
Overview
CompanyMistralNVIDIA
Release dateAug 12 2025Jun 4 2026
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Nemotron 3 Ultra leads Mistral Medium 3.1 on 1 of the 1 benchmark they both report (GPQA Diamond). Mistral Medium 3.1 shipped 296 days before Nemotron 3 Ultra, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Mistral Medium 3.1) vs 1M (Nemotron 3 Ultra). Mistral Medium 3.1 is closed, while Nemotron 3 Ultra is open weight.

On GPQA Diamond, Nemotron 3 Ultra leads at 86.7% vs Mistral Medium 3.1 at 65.7%.