Compare AI models

Specifications
Parameters
671B24B
Context window
128k128k
API pricing
Cheapest input
$0.25DeepInfra—
Cheapest output
$0.95DeepInfra—
$97.50DeepInfra · fp4—
Benchmarks
LiveCodeBench
74.8%70.88%
AIME 2024
93.1%86.14%
AIME 2025
88.4%77.34%
GPQA Diamond
80.1%70.07%
Overview
CompanyDeepSeekMistral
Release dateAug 21 2025Sep 17 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek V3.1 leads Magistral Small 1.2 on 4 of the 4 benchmarks they both report (LiveCodeBench, AIME 2024, AIME 2025, GPQA Diamond). DeepSeek V3.1 shipped 27 days before Magistral Small 1.2, so benchmark comparisons should account for the intervening progress.

DeepSeek V3.1 has 671B parameters, while Magistral Small 1.2 has 24B. Context windows are 128k (DeepSeek V3.1) vs 128k (Magistral Small 1.2).

On LiveCodeBench, DeepSeek V3.1 leads at 74.8% vs Magistral Small 1.2 at 70.88%. On AIME 2024, DeepSeek V3.1 leads at 93.1% vs Magistral Small 1.2 at 86.14%. On AIME 2025, DeepSeek V3.1 leads at 88.4% vs Magistral Small 1.2 at 77.34%. On GPQA Diamond, DeepSeek V3.1 leads at 80.1% vs Magistral Small 1.2 at 70.07%.