AIME 2025

The 30 problems from the 2025 American Invitational Mathematics Examination, scored as an exact-answer test of competition-level math reasoning.

Also known as: AIME25, AIME 2025 I and II

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categorymath
Subcategorycompetition mathematics
Page statusactive
Metricaccuracy (pass@1, exact match on the final integer)
Directionhigher_is_better
Unit%
Dataset size30
Dataset licenceApache-2.0
PublisherMathematical Association of America (MAA)

What it measures

AIME 2025 gives a model the problems from the 2025 American Invitational Mathematics Examination, a competition that only the top-scoring AMC participants sit, and checks whether the model returns the single correct integer answer. It exercises multi-step algebra, geometry, number theory and combinatorics reasoning well above the level of grade-school math benchmarks, and gives no partial credit for a sound method that lands on the wrong final number.

Task format

Free-response competition math problem in, single integer answer from 0 to 999 out; no answer choices are offered.

Models reporting this benchmark

These figures come from the model cards, which carry one collection date per card and no per-score attribution. They are shown as reported, not as verified evidence.
ModelProviderScoreCard as of
GLM 5.1Z.ai (Zhipu AI)95.32026-04
o3-proOpenAI942026-04
o3OpenAI91.52026-04
o3-deep-researchOpenAI91.52026-04
o4-miniOpenAI87.02026-04
o4-mini-deep-researchOpenAI87.02026-04
Gemini 2.5 ProGoogle DeepMind86.72026-04
Gemini 2.5 Pro Preview 05-06Google DeepMind86.72026-04
Gemini 2.5 Pro Preview 06-05Google DeepMind86.72026-04
Gemini 2.5 Pro Preview TTSGoogle DeepMind86.72026-04
o3-miniOpenAI83.32026-04
DeepSeek R1 0528DeepSeek82.52026-04
DeepSeek R1 0528 NVFP4 v2NVIDIA82.52026-04
Claude Opus 4Anthropic82.02026-04
Claude Opus 4.6Anthropic82.02026-04
DeepSeek R1DeepSeek79.82026-04
DeepSeek ReasonerDeepSeek79.82026-04
Claude Opus 4.5Anthropic78.02026-04
Claude Opus 4.5 (latest)Anthropic78.02026-04
Gemini 2.5 FlashGoogle DeepMind73.32026-04
Gemini 2.5 Flash ImageGoogle DeepMind73.32026-04
Gemini 2.5 Flash Image (Preview)Google DeepMind73.32026-04
Gemini 2.5 Flash LiteGoogle DeepMind73.32026-04
Gemini 2.5 Flash Lite Preview 06-17Google DeepMind73.32026-04
Gemini 2.5 Flash Lite Preview 09-25Google DeepMind73.32026-04
Gemini 2.5 Flash Preview 04-17Google DeepMind73.32026-04
Gemini 2.5 Flash Preview 05-20Google DeepMind73.32026-04
Gemini 2.5 Flash Preview 09-25Google DeepMind73.32026-04
Gemini 2.5 Flash Preview TTSGoogle DeepMind73.32026-04

Data

This page as JSON · Edit on GitHub