American Invitational Mathematics Examination 2025

The most recent AIME examination, featuring 15 challenging mathematics problems testing olympiad-level mathematical reasoning with integer answers from 000-999.

In index
Math
Best in math

Models53

Top result100.0%GPT-5.2

Top-3 spread2.7 pts

Year—

Result and price

24 of 53 results

Results

53 results

#ModelResult

01GPT-5.2100.0%

02Ripple-65B Preview98.3%

03Step 3.5 Flash97.3%

04MAI-Thinking-197.0%

05OrcaCyber-Nano-1.096.2%

06Kimi K2.5 (Reasoning)96.1%

06Kimi K2.596.1%

08GLM-4.795.7%

09Ternary Bonsai 2 27B95.0%

10MiMo-V2-Flash94.1%

11DeepSeek V3.2 (Thinking)93.1%

12o4-mini (high)92.7%

13GPT-OSS 120B92.5%

14Qwen3 235B 2507 (Reasoning)92.3%

15GPT-OSS 20B91.7%

16GLM-4.7-Flash91.6%

17Nemotron 3 Super 100B90.2%

18Granite 4.2 30B89.2%

19Nemotron 3 Nano 30B89.1%

20DeepSeek V3.1 (Reasoning)88.4%

21Sarvam 105B88.3%

22DeepSeek-R1-052887.5%

23Claude Sonnet 4.587.0%

24Granite 4.2 8B86.7%

25MiniCPM5-2B86.5%

26Exaone 4.0 32B85.3%

27Ministral 3 14B (Reasoning)85.0%

28Nemotron 3 Nano Omni 30B A3B82.1%

29Qwen3 235B-A22B (Reasoning)81.5%

30Granite 4.2 3B78.3%

31Phi-4-reasoning-plus78.0%

32MiniMax M1 80k76.9%

33Claude 4 Opus Thinking75.5%

34Qwen3 32B (Reasoning)72.9%

35Nemotron Ultra 253B72.5%

36Qwen3 235B 250770.3%

37QwQ-32B69.5%

38Magistral Medium64.9%

39Phi-4-reasoning62.9%

40Magistral Small62.8%

41LFM2.5-2.6B51.9%

42DeepSeek V3.149.8%

42Gemini 2.5 Flash-Lite49.8%

44Kimi K249.5%

45LFM2.5-8B-A1B42.5%

46MiniCPM5-1B40.4%

47Claude 4 Opus33.9%

48Claude 4 Sonnet33.1%

49Gemini 2.0 Flash29.7%

50Qwen3 235B-A22B24.7%

51Gemini 2.0 Flash-Lite23.8%

52Qwen3 32B20.2%

53Gemini 1.5 Flash14.7%