Compare AI models

Specifications
Parameters
671B24B
Context window
128k40k
Benchmarks
LiveCodeBench
88.7%55.84%
AIME 2025
96%62.76%
GPQA Diamond
85.7%68.18%
Overview
CompanyDeepSeekMistral
Release dateDec 1 2025Jun 10 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-V3.2-Speciale leads Magistral Small on 3 of the 3 benchmarks they both report (LiveCodeBench, AIME 2025, GPQA Diamond). Magistral Small shipped 174 days before DeepSeek-V3.2-Speciale, so benchmark comparisons should account for the intervening progress.

DeepSeek-V3.2-Speciale has 671B parameters, while Magistral Small has 24B. Context windows are 128k (DeepSeek-V3.2-Speciale) vs 40k (Magistral Small).

On LiveCodeBench, DeepSeek-V3.2-Speciale leads at 88.7% vs Magistral Small at 55.84%. On AIME 2025, DeepSeek-V3.2-Speciale leads at 96% vs Magistral Small at 62.76%. On GPQA Diamond, DeepSeek-V3.2-Speciale leads at 85.7% vs Magistral Small at 68.18%.