MATH

Math% accuracy

12,500 competition math problems across 7 subjects.

Updated Jun 1, 2025Latest measured Jun 1, 20250 verified · 9 self-reported

Verified results come from third-party or public leaderboard sources. Self-reported results come from provider papers, blogs, or vendor disclosures and should be compared with extra caution.

At a glance

🏆 Top score

o3 OpenAI97.8 % accuracy

Total results

Models tested

Providers

Verified · Self-reported

0 · 9

Average

87.68 % accuracy

Median

90 % accuracy

Range

70.2 – 97.8 % accuracy

Score distribution

Methodology

Numerical and closed-form answers on problems from AMC/AIME-style sources.

Limitations

Some training contamination on older models. New problems needed over time.

By provider

OpenAI· 5 models
97.8 % accuracy
o3
Average: 87.98 % accuracyBest: 97.8 % accuracy
DeepSeek· 1 model
97.3 % accuracy
DeepSeek R1
Average: 97.3 % accuracyBest: 97.3 % accuracy
xAI· 1 model
93.3 % accuracy
Grok 3

Full leaderboard

Showing 9 of 9

ProviderSourceSort by

#	Model	Provider	Score (% accuracy)	Source	Date
1	o3	OpenAI	97.8	Self-reported OpenAI o3 system card	Jun 1, 2025
2	DeepSeek R1	DeepSeek	97.3	Self-reported DeepSeek R1 paper	Jun 1, 2025
3	o1	OpenAI	94.8	Self-reported OpenAI o1 system card	Jun 1, 2025
4	Grok 3	xAI	93.3	Self-reported xAI model card	Jun 1, 2025
5	o3-mini	OpenAI	90	Self-reported OpenAI system card	Jun 1, 2025
6	GPT-5	OpenAI	87.1	Self-reported OpenAI system card	Jun 1, 2025
7	Phi-4	Microsoft	80.4	Self-reported Microsoft Phi-4 tech report	Jun 1, 2025
8	Claude Opus 4	Anthropic	78.2	Self-reported Anthropic model card	Jun 1, 2025
9	GPT-4o mini	OpenAI	70.2	Self-reported OpenAI system card	Jun 1, 2025

Community ratings

No ratings yet. Be the first to rate MATH.