Compare AI models

Specifications
Parameters
671B—
Context window
128k128k
Benchmarks
LiveCodeBench
88.7%74.8%
AIME 2025
96%83%
GPQA Diamond
85.7%79%
Overview
CompanyDeepSeekSpaceXSI
Release dateDec 1 2025Feb 17 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek-V3.2-Speciale leads Grok 3 mini on 3 of the 3 benchmarks they both report (LiveCodeBench, AIME 2025, GPQA Diamond). Grok 3 mini shipped 287 days before DeepSeek-V3.2-Speciale, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek-V3.2-Speciale) vs 128k (Grok 3 mini). DeepSeek-V3.2-Speciale is open weight, while Grok 3 mini is closed.

On LiveCodeBench, DeepSeek-V3.2-Speciale leads at 88.7% vs Grok 3 mini at 74.8%. On AIME 2025, DeepSeek-V3.2-Speciale leads at 96% vs Grok 3 mini at 83%. On GPQA Diamond, DeepSeek-V3.2-Speciale leads at 85.7% vs Grok 3 mini at 79%.