AIME 2025 Leaderboard

American Invitational Mathematics Examination 2025 problems testing olympiad-level mathematical reasoning.

Data from Artificial Analysis

As of November 21, 2026, the top-scoring model on AIME 2025 is GPT-5.2 Pro at 99.0%, followed by GPT-5 Codex at 98.7% and Gemini 3 Flash Preview at 97.0%. 170 models have been evaluated on this benchmark.

Last updated: November 21, 2026

8 Ways to Use Fewer Tokens

Models

170

Best Score

99.0

Average

53.0

Std Dev

30.9

Categories
Mathematical Problem Solving
Sponsored The Grid | Spot Priced LLM API
The Grid | Spot Priced LLM API — Why pay list price when providers will bid for your API usage? Up to 80% off.
View spot prices
Provider
Model
Input $/M
Output $/M
AIME 2025
Actions
$10.500
$84.000
99.0
$1.250
$10.000
98.7
$0.500
$3.000
97.0
$0.270
$0.400
96.7
$1.750
$14.000
96.7
$0.090
$0.290
96.3
$1.250
$10.000
95.7
$2.000
$12.000
95.7
$0.320
$1.400
95.0
$0.550
$2.200
94.7
$0.207
$0.828
94.7
$1.250
$10.000
94.3
$0.625
$5.000
94.0
$0.030
$0.100
93.4
$3.000
$15.000
92.7
$0.209
$0.310
92.0
$0.250
$2.000
91.7
$5.000
$25.000
91.3
$0.450
$0.900
91.0
$0.060
$0.240
91.0
$1.100
$4.400
90.7
$0.125
$1.000
90.7
$0.250
$0.950
89.7
$0.200
$0.500
89.7
$0.270
$1.000
89.7
$0.017
$0.085
89.3
$0.000
$0.000
89.3
$2.000
$8.000
88.3
$0.900
$0.900
88.3
$3.000
$15.000
88.0
$0.200
$1.100
88.0
$1.000
$10.000
87.7
$0.270
$0.410
87.7
$0.430
$1.750
86.0
$0.300
$0.900
85.3
$0.250
$0.500
84.7
$0.500
$0.900
84.7
$0.150
$0.900
84.3
$0.025
$0.200
83.7
$1.000
$5.000
83.7
$0.200
$0.200
82.7
$0.300
$1.200
82.7
$0.900
$0.900
82.3
$0.780
$3.900
82.3
$0.900
$0.900
82.0
$0.130
$0.850
80.7
$0.780
$3.900
80.7
$15.000
$75.000
80.3
$0.300
$1.200
78.3
$0.150
$0.500
77.3
$0.000
$0.000
76.7
$0.500
$2.150
76.0
$0.200
$0.200
75.0
$3.000
$15.000
74.3
$0.600
$2.200
73.7
$0.150
$0.500
73.7
$15.000
$75.000
73.3
$0.300
$2.500
73.3
$0.064
$0.224
73.0
$0.600
$1.800
73.0
$0.900
$0.900
72.7
$0.120
$0.500
72.3
$0.150
$0.600
72.3
$0.000
$0.000
71.7
$0.200
$0.880
70.7
$0.120
$0.200
70.7
$1.000
$3.000
69.7
$0.060
$0.200
69.7
$0.093
$0.373
68.7
$0.100
$0.400
68.7
$0.500
$0.900
68.3
$0.550
$2.190
68.0
$0.090
$0.290
67.7
$0.090
$0.300
66.3
$0.090
$0.900
66.3
$0.300
$0.300
63.0
$5.000
$25.000
62.7
$0.060
$0.200
62.3
$0.000
$0.000
61.0
$0.300
$2.500
60.3
$0.209
$0.310
59.0
$0.080
$0.200
58.0
$3.000
$15.000
58.0
$0.270
$0.410
57.7
$0.600
$2.500
57.3
$3.000
$15.000
56.3
$0.200
$0.900
56.3
$0.080
$0.200
55.7
$0.500
$3.000
55.7
$0.750
$0.800
53.7
$0.270
$1.000
53.7
$0.100
$0.400
53.3
$0.200
$0.200
52.3
$0.250
$0.950
49.7
$1.250
$10.000
48.3
$0.320
$1.400
48.0
$0.100
$0.400
46.7
$0.400
$1.600
46.3
$0.430
$1.750
44.3
$0.200
$1.500
43.3
$0.900
$0.900
41.3
$0.200
$0.500
41.3
$0.100
$0.200
41.3
$0.250
$1.000
41.0
$0.300
$0.900
39.3
$1.000
$5.000
39.0
$0.440
$2.200
38.3
$3.000
$15.000
38.0
$3.000
$15.000
37.0
$0.400
$0.900
36.7
$15.000
$75.000
36.3
$0.100
$0.400
35.3
$2.000
$8.000
34.7
$0.000
$0.000
34.3
$0.300
$2.500
33.7
$0.150
$0.150
31.7
$0.117
$1.365
30.7
$0.440
$2.200
30.3
$0.200
$0.200
30.0
$0.070
$0.280
29.3
$0.900
$0.900
29.0
$0.070
$0.280
29.0
$0.180
$0.200
27.3
$0.075
$0.200
27.0
$0.200
$0.200
26.7
$0.300
$0.900
26.3
$0.010
$0.020
25.3
$0.200
$0.200
24.3
$0.100
$0.400
24.0
$0.900
$0.900
23.7
$0.150
$0.600
21.7
$0.120
$0.500
21.7
$3.000
$15.000
21.0
$0.080
$0.160
20.7
$0.064
$0.224
19.7
$0.200
$0.600
19.3
$0.200
$0.200
19.0
$0.050
$0.150
18.3
$0.070
$0.140
18.0
$2.500
$12.500
17.3
$0.600
$1.800
15.3
$1.000
$3.000
15.3
$0.150
$0.600
14.7
$0.000
$0.000
14.3
$0.288
$0.320
14.0
$2.000
$6.000
14.0
$0.100
$0.300
14.0
$0.060
$0.240
13.3
$2.500
$10.000
13.0
$0.040
$0.050
12.7
$0.093
$0.373
11.3
$0.880
$0.880
11.0
$0.000
$0.000
8.0
$0.100
$0.320
7.7
$0.800
$3.200
7.0
$0.060
$0.240
7.0
$0.035
$0.140
6.0
$0.400
$2.000
4.7
$0.020
$0.040
4.3
$0.040
$0.064
4.3
$0.340
$0.390
4.0
$0.100
$0.300
3.7
$0.030
$0.050
3.3
$0.050
$0.200
3.3
$1.500
$1.500
3.0
$2.000
$6.000
2.3
$2.000
$8.000
2.3
$0.060
$0.060
1.7
$0.020
$0.020
-
$2.000
$6.000
-

Pricing from OpenRouter. Benchmarks from Artificial Analysis.

About AIME 2025

American Invitational Mathematics Examination 2025 problems testing olympiad-level mathematical reasoning.

This leaderboard shows all models with AIME 2025 benchmark scores, ranked from highest to lowest. Pricing data is included to help you compare performance against cost.

Frequently Asked Questions

American Invitational Mathematics Examination 2025 problems testing olympiad-level mathematical reasoning.
As of November 21, 2026, GPT-5.2 Pro leads the AIME 2025 leaderboard with a score of 99.0. Rankings change as new models are released and evaluated.
Currently 170 models have been evaluated on AIME 2025, with an average score of 53.0 and standard deviation of 30.9.
Benchmark scores are updated when new evaluations are published by our data sources (Artificial Analysis and LayerLens). Pricing data is refreshed daily from OpenRouter.