Compare AI models

Specifications
Parameters
1.5B/7B/8B/14B/32B/70B—
Context window
—1M
API pricing
Cheapest input
—$0.10Azure
Cheapest output
—$0.40Azure
—$40.00Azure
Benchmarks
MATH-500
94.5%62.3%
AIME 2024
70%29.4%
GPQA Diamond
65.2%50.3%
Overview
CompanyDeepSeekOpenAI
Release dateJan 20 2025Apr 14 2025
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

DeepSeek R1 Distill leads GPT-4.1 nano on 3 of the 3 benchmarks they both report (MATH-500, AIME 2024, GPQA Diamond). DeepSeek R1 Distill shipped 84 days before GPT-4.1 nano, so benchmark comparisons should account for the intervening progress.

DeepSeek R1 Distill is open weight, while GPT-4.1 nano is closed.

On MATH-500, DeepSeek R1 Distill leads at 94.5% vs GPT-4.1 nano at 62.3%. On AIME 2024, DeepSeek R1 Distill leads at 70% vs GPT-4.1 nano at 29.4%. On GPQA Diamond, DeepSeek R1 Distill leads at 65.2% vs GPT-4.1 nano at 50.3%.