AIME 2025

American Invitational Mathematics Examination 2025 (I and II)

The 30 problems from the 2025 AIME I (6 February 2025) and AIME II (12 February 2025), each with an integer answer from 000 to 999. Scored as the fraction of problems answered exactly, usually averaged over several samples (mean pass@1). Because the exam is written fresh each year, it was a clean test of competition-math reasoning for models trained before February 2025; scores with a code interpreter are not comparable to no-tools runs.

website · leaderboard · code

100%
GPT-5.2 (high)
independent
2025-02-12 → 2025-12-11: 70% → 100%
Released
2025-02
Maintainer
Mathematical Association of America (exam); evaluated by MathArena and model developers
Status
saturated
Contamination
high
Metric
accuracy (percent, ↑)
Tasks
30
Domains
math reasoning
human
no measured baseline

Notes. Scores are almost always reported on I+II combined (30 problems); a single exam has 15. Problems and solutions have been public since February 2025, so models released later have likely seen them. Several frontier systems report 100% (with or without tools), so the exam no longer separates them; MathArena marks AIME 2025 as deprecated in favour of newer competitions.

Full ledger

SystemDeveloperScoreDateSourceConditions
GPT-5.2 (high)OpenAI100%independenttools: no pass_k: 1
MathArena AIME 2025 table, 100.00% +/- 0.00 for both high and xhigh effort. Date is model release; MathArena marks the competition deprecated.
GPT-5 (no tools)OpenAI94.6%self-reportedtools: no
With thinking. GPT-5 with python scored 99.6 in the same chart; tool runs are not comparable.
o3 (high)OpenAI88.9%self-reportedtools: no
GPT-5 launch chart 'AIME 2025 (no tools)'; MathArena independently measured o3 (high) at 89.2.
DeepSeek-R1DeepSeek70%independenttools: no pass_k: 1
MathArena AIME 2025 table (I+II, 30 problems), 70.00% +/- 8.20. MathArena does not show a run date; date used is AIME II exam day (R1 was released 2025-01-20, before the exam).