HyperAI
Command Palette
Search for a command to run...
AIME 2025
2025 American Invitational Mathematics Examination의 30개 문제 전체로, 000-999의 정수 답안을 사용하는 올림피아드 수준의 수학적 추론을 시험합니다.
- 모델
- 78 현재, 270 전체
- 중앙값
- 51.3%
- 범위
- 0.0%–94.3%
| 순위 | 비교 | 모델 | 점수 |
|---|---|---|---|
| 1 |  Nova 2.0 Lite (High) Amazon | 94.3% | |
| 2 |  gpt-oss-120b (High) OpenAI | 93.4% | |
| 3 | 91.0% | ||
| 4 | Nova 2.0 Omni (Medium) Amazon | 89.7% | |
| 5 | gpt-oss-20b (High) OpenAI | 89.3% | |
| 6 | 89.0% | ||
| 7 | Nova 2.0 Lite (Medium) Amazon | 88.7% | |
| 8 | o3 OpenAI | 88.3% | |
| 9 |  Apriel-v1.6-15B-Thinker ServiceNow | 88.0% | |
| 10 |  INTELLECT-3 (Based on GLM-4.5-Air) Prime Intellect | 88.0% | |
| 11 | 85.0% | ||
| 12 | 84.3% | ||
| 13 |  Ring-flash-2.0 InclusionAI | 83.7% | |
| 14 |  Magistral Medium 1.2 Mistral | 82.0% | |
| 15 | Magistral Small 1.2 Mistral | 80.3% | |
| 16 |  Motif-2-12.7B-Reasoning Motif Technologies | 80.3% | |
| 17 |  EXAONE 4.0 32B (Reasoning) LG AI Research | 80.0% | |
| 18 |  Falcon-H1R-7B TII UAE | 80.0% | |
| 19 |  Doubao Seed Code ByteDance Seed | 79.3% | |
| 20 |  K2-V2 (High) Institute of Foundation Models | 78.3% | |
| 21 |  Olmo 3.1 32B Think Allen Institute for AI | 77.3% | |
| 22 |  Mi:dm K 2.5 Pro Korea Telecom | 76.7% | |
| 23 | 75.0% | ||
| 24 | 74.0% | ||
| 25 |  Cogito v2.1 (Reasoning) Deep Cogito | 72.7% | |
| 26 | Olmo 3 7B Think Allen Institute for AI | 70.7% | |
| 27 |  Hermes 4 - Llama-3.1 405B (Reasoning) Nous Research | 69.7% | |
| 28 | 69.7% | ||
| 29 | Hermes 4 - Llama-3.1 70B (Reasoning) Nous Research | 68.7% | |
| 30 | gpt-oss-120b (Low) OpenAI | 66.7% | |
| 31 |  Qwen3 Next 80B A3B Instruct Alibaba | 66.3% | |
| 32 | K2-V2 (Medium) Institute of Foundation Models | 64.7% | |
| 33 | 63.7% | ||
| 34 | 63.3% | ||
| 35 | gpt-oss-20b (Low) OpenAI | 62.3% | |
| 36 | 62.3% | ||
| 37 | 59.0% | ||
| 38 | Nova 2.0 Omni (Low) Amazon | 56.0% | |
| 39 | Qwen3 Omni 30B A3B Instruct Alibaba | 52.3% | |
| 40 | Exaone 4.0 1.2B (Reasoning) LG AI Research | 50.3% | |
| 41 | Nova 2.0 Lite (Low) Amazon | 46.7% | |
| 42 |  ERNIE 4.5 300B A47B Baidu | 41.3% | |
| 43 | Olmo 3 7B Instruct Allen Institute for AI | 41.3% | |
| 44 | EXAONE 4.0 32B (Non-reasoning) LG AI Research | 39.3% | |
| 45 | 37.0% | ||
| 46 | 36.3% | ||
| 47 | K2-V2 (Low) Institute of Foundation Models | 35.3% | |
| 48 | 33.7% | ||
| 49 |  Reka Flash 3 Reka AI | 33.7% | |
| 50 | Ministral 3 8B Mistral | 31.7% |
