Compare AI models

Specifications
Parameters
—671B
Context window
200k128k
Benchmarks
GPQA Diamond
79.6%85.7%
Overview
CompanyAnthropicDeepSeek
Release dateMay 22 2025Dec 1 2025
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-V3.2-Speciale leads Claude Opus 4 on 1 of the 1 benchmark they both report (GPQA Diamond). Claude Opus 4 shipped 193 days before DeepSeek-V3.2-Speciale, so benchmark comparisons should account for the intervening progress.

Context windows are 200k (Claude Opus 4) vs 128k (DeepSeek-V3.2-Speciale). Claude Opus 4 is closed, while DeepSeek-V3.2-Speciale is open weight.

On GPQA Diamond, DeepSeek-V3.2-Speciale leads at 85.7% vs Claude Opus 4 at 79.6%.