Compare AI models

Specifications
Parameters
671B—
Context window
128k2M
Benchmarks
LiveCodeBench
65.9%34.3%
GPQA Diamond
71.5%59.1%
MMLU-Pro
84%75.8%
Overview
CompanyDeepSeekGoogle
Release dateJan 20 2025Sep 24 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek R1 leads Gemini 1.5 Pro-002 on 3 of the 3 benchmarks they both report (LiveCodeBench, GPQA Diamond, MMLU-Pro). Gemini 1.5 Pro-002 shipped 118 days before DeepSeek R1, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek R1) vs 2M (Gemini 1.5 Pro-002). DeepSeek R1 is open weight, while Gemini 1.5 Pro-002 is closed.

On LiveCodeBench, DeepSeek R1 leads at 65.9% vs Gemini 1.5 Pro-002 at 34.3%. On GPQA Diamond, DeepSeek R1 leads at 71.5% vs Gemini 1.5 Pro-002 at 59.1%. On MMLU-Pro, DeepSeek R1 leads at 84% vs Gemini 1.5 Pro-002 at 75.8%.