The 2024 American Invitational Mathematics Examination, consisting of 15 challenging problems completed in a three-hour session.
Browse the latest scores, model modes, release dates, and parameter sizes for AIME 2024.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
| Rank | Model | License | |||
|---|---|---|---|---|---|
 OpenAI o4 - mini Thinking EnabledTools | 98.70 | 2025-04-16 | Unknown | Closed | |
GPT OSS 120B Thinking EnabledTools | 96.60 | 2025-08-06 | 11.7B | Free Commercial | |
 Grok-3 mini - Reasoning Standard Mode | 96.00 | 2025-02-18 | Unknown | Closed | |
4 | GPT OSS 20B Thinking EnabledTools | 96.00 | 2025-08-06 | 21B | Free Commercial |
5 | OpenAI o4 - mini Thinking Enabled | 93.40 | 2025-04-16 | Unknown | Closed |
6 | Grok-3 - Reasoning Beta Standard Mode | 93.30 | 2025-02-18 | Unknown | Closed |
7 |  DeepSeek-V3.1 Thinking Enabled | 93.10 | 2025-08-20 | 671B | Free Commercial |
8 | o3-pro Standard Mode | 93.00 | 2025-06-10 | Unknown | Closed |
9 |  Gemini 2.5 Pro Experimental 03-25 Standard Mode | 92.00 | 2025-03-25 | Unknown | Closed |
10 | Gemini-2.5-Pro-Preview-05-06 Standard Mode | 92.00 | 2025-05-06 | Unknown | Closed |
11 | Gemini 2.5-Pro Standard Mode | 92.00 | 2025-06-05 | Unknown | Closed |
12 | OpenAI o3 Standard Mode | 91.60 | 2025-04-16 | Unknown | Closed |
13 | DeepSeek-R1-0528 Thinking Enabled | 91.40 | 2025-05-28 | 671B | Free Commercial |
14 |  GLM-4.5 Thinking Enabled | 91.00 | 2025-07-28 | 355B | Free Commercial |
15 | GLM-4.5-Air Thinking Enabled | 89.40 | 2025-07-28 | 106B | Free Commercial |
16 | Gemini 2.5 Flash Standard Mode | 88.00 | 2025-04-17 | Unknown | Closed |
17 |  Hunyuan-A13B-Instruct Standard Mode | 87.30 | 2025-06-27 | 80B | Free Commercial |
18 | OpenAI o3-mini (high) Standard Mode | 87.00 | 2025-01-31 | Unknown | Closed |
19 |  MiniMax-M1-80k Standard Mode | 86.00 | 2025-06-16 | 456B | Free Commercial |
20 |  Qwen3-235B-A22B Standard Mode | 85.70 | 2025-04-28 | 235B | Free Commercial |
21 | Qwen3-235B-A22B Thinking Enabled | 85.70 | 2025-04-28 | 235B | Free Commercial |
22 | Grok 3 Standard Mode | 84.20 | 2025-02-17 | Unknown | Closed |
23 | MiniMax-M1-40k Standard Mode | 83.30 | 2025-06-16 | 456B | Free Commercial |
24 |  Pangu Embedded Standard Mode | 81.90 | 2025-06-30 | 7B | Free Commercial |
25 | Qwen3-32B Thinking Enabled | 81.40 | 2025-04-28 | 32B | Free Commercial |
26 |  Hunyuan-7B Standard Mode | 81.10 | 2025-08-04 | 7B | Free Commercial |
27 |  Claude Sonnet 3.7-64K Extended Thinking Standard Mode | 80.00 | 2025-02-25 | Unknown | Closed |
28 | DeepSeek-R1 Standard Mode | 79.80 | 2025-01-20 | 671B | Free Commercial |
29 | QwQ-32B Standard Mode | 79.50 | 2025-03-06 | 32.5B | Free Commercial |
30 | Qwen3-8B Standard Mode | 79.40 | 2025-04-28 | 8B | Free Commercial |
