Compare AI models

Specifications
Parameters
671B—
Context window
128k8k
API pricing
Cheapest input
—$30.00Azure
Cheapest output
—$60.00Azure
—$9,000Azure
Benchmarks
DROP
92.280.9
MMLU
90.8%86.4%
Overview
CompanyDeepSeekOpenAI
Release dateJan 20 2025Mar 14 2023
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek R1 leads GPT-4 on 2 of the 2 benchmarks they both report (DROP, MMLU). GPT-4 shipped 678 days before DeepSeek R1, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek R1) vs 8k (GPT-4). DeepSeek R1 is open weight, while GPT-4 is closed.

On DROP, DeepSeek R1 leads at 92.2 vs GPT-4 at 80.9. On MMLU, DeepSeek R1 leads at 90.8% vs GPT-4 at 86.4%.