Compare AI models

Specifications
Parameters
671B—
Context window
128k128k
Benchmarks
MATH-500
97.3%90%
DROP
92.283.9
AIME 2024
79.8%70%
GPQA Diamond
71.5%60%
MMLU
90.8%85.2%
Overview
CompanyDeepSeekOpenAI
Release dateJan 20 2025Sep 12 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek R1 leads o1-mini on 5 of the 5 benchmarks they both report. o1-mini shipped 130 days before DeepSeek R1, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek R1) vs 128k (o1-mini). DeepSeek R1 is open weight, while o1-mini is closed.

On MATH-500, DeepSeek R1 leads at 97.3% vs o1-mini at 90%. On DROP, DeepSeek R1 leads at 92.2 vs o1-mini at 83.9. On AIME 2024, DeepSeek R1 leads at 79.8% vs o1-mini at 70%. On GPQA Diamond, DeepSeek R1 leads at 71.5% vs o1-mini at 60%. On MMLU, DeepSeek R1 leads at 90.8% vs o1-mini at 85.2%.