Compare AI models

Specifications
Parameters
24B—
Context window
128k128k
API pricing
Cheapest input
$0.06DeepInfra—
Cheapest output
$0.16DeepInfra—
$25.00DeepInfra · fp8—
Benchmarks
MATH
69.42%57.9%
MMLU
80.5%81.9%
Overview
CompanyMistralZ.ai
Release dateJun 20 2025Jun 5 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Mistral Small 3.2 and GLM-4-Air are evenly matched across the 2 benchmarks they both report (MATH, MMLU). GLM-4-Air shipped 380 days before Mistral Small 3.2, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Mistral Small 3.2) vs 128k (GLM-4-Air). Mistral Small 3.2 is open weight, while GLM-4-Air is closed.

On MATH, Mistral Small 3.2 leads at 69.42% vs GLM-4-Air at 57.9%. On MMLU, GLM-4-Air leads at 81.9% vs Mistral Small 3.2 at 80.5%.