Compare AI models

Specifications
Parameters
—16B
Context window
128k8k
Benchmarks
GPQA Diamond
65.7%35.2%
MMLU-Pro
76.8%42.4%
Overview
CompanyMistralMoonshot AI
Release dateAug 12 2025Feb 22 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Mistral Medium 3.1 leads Moonlight on 2 of the 2 benchmarks they both report (GPQA Diamond, MMLU-Pro). Moonlight shipped 171 days before Mistral Medium 3.1, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Mistral Medium 3.1) vs 8k (Moonlight). Mistral Medium 3.1 is closed, while Moonlight is open weight.

On GPQA Diamond, Mistral Medium 3.1 leads at 65.7% vs Moonlight at 35.2%. On MMLU-Pro, Mistral Medium 3.1 leads at 76.8% vs Moonlight at 42.4%.