Inference
EN
Vals AI

AIME

Challenging national math exam given to top high-school students

ModelsModels with a result.94
Top resultBest result on this benchmark.98.1%Gemini 3.1 Pro Preview
Top-3 spreadPoints from first to third.1.3 pts
UpdatedDate the source changed the results.16 Apr 2026

Result and cost

91 of 95 results

Results

95 results
#ModelResultThe score from the source.CostUS dollars to run one task.TimeTime to run one task.
01Gemini 3.1 Pro Preview98.1% ±0.3$0.1272 min
02Muse Spark96.9% ±0.4$0.00463 min
02GPT-5.296.9% ±0.2$0.1352 min
04Gemini 3 Pro96.7% ±0.4$0.1692 min
04GPT-5.496.7% ±0.5$0.1382 min
06Grok 4.20 (Reasoning)96.5% ±0.5$0.0592 min
07Claude Opus 4.796.3% ±0.5$0.4513 min
08Claude Opus 4.6 (Adaptive)95.6% ±0.6$0.3445 min
08Gemini 3 Flash Preview95.6% ±0.5$0.04684 s
08GPT-5.4 mini95.6% ±0.4$0.1295 min
08Kimi K2.5 Thinking95.6% ±0.3$0.0645 min
12Claude Opus 4.5 Thinking95.4% ±0.4$1.324 min
13Qwen3.6 Plus94.6% ±0.6$0.0949 min
14GPT-593.4% ±1.3$0.0935 min
15GPT-5.193.3% ±0.6$0.153 min
15GLM-4.793.3% ±0.4$0.0439 min
17GLM-4.692.7% ±0.6$0.11612 min
18GPT-OSS 120B92.6% ±1.2$0.013 min
19Qwen3.5 Flash92.5% ±0.9$0.0114 min
20Claude Sonnet 4.692.3% ±0.4$0.4365 min
21GLM-5.191.9% ±0.4$0.0716 min
21Grok 4.1 Fast (Reasoning)91.9% ±0.6$0.00412 min
23GLM-5 Thinking91.7% ±0.8$0.078 min
24GPT-5 mini91.5% ±0.8$0.0232 min
25Grok 4 Fast (Reasoning)91.3% ±1.0$0.004335 s
26MiniMax M2.791.0% ±0.7$0.0234 min
27Grok 490.6% ±1.8$0.1742 min
28GPT-5.4 nano88.8% ±1.0$0.003466 s
28MiniMax M2.588.8% ±0.9$0.0416 min
30Claude Sonnet 4.5 Thinking88.2% ±0.7$0.2674 min
31GLM-4.586.7% ±1.3$0.0476 min
32o3-mini86.5% ±1.1$0.0513 min
33Qwen3.5 Plus Thinking86.0% ±1.1$0.0454 min
33GPT-OSS 20B86.0% ±1.1$0.0114 min
35Gemini 2.5 Pro Exp 03.2585.8% ±1.3$0.0272 min
36Kimi K2 Thinking85.4% ±1.3$0.04515 min
37o385.3% ±1.4$0.064 min
38Grok 3 Mini Fast High Reasoning85.0% ±0.5$0.00832 min
39DeepSeek V3.2 (Thinking)84.6% ±1.1$0.00446 min
40Qwen3 235B-A22B84.0% ±0.8$0.0124 min
41o4-mini83.7% ±1.2$0.02655 s
42Magistral MediumAbove fitFar above this model's index.83.5% ±1.0$0.0877 min
43Gemini 3.1 Flash Lite Preview83.3% ±1.2$0.00741 s
44Claude Haiku 4.5 Thinking82.7% ±0.9$0.1743 min
45GPT-5 nano81.2% ±1.3$0.00893 min
46Qwen3 Maxqwen3-max81.0% ±1.1$0.03317 s
47Magistral SmallAbove fitFar above this model's index.80.7% ±1.2$0.0254 min
48Claude Opus 4.1 Thinking78.2% ±1.6$0.864 min
49MiniMax M2.177.9% ±1.3$0.0244 min
50Claude Opus 4.576.9% ±1.4$0.08917 s
51Claude Sonnet 4 Thinking76.3% ±1.4$0.2955 min
52DeepSeek-R174.0% ±1.3$0.0533 min
53o171.5% ±1.5$0.6593 min
54Grok 3 Mini Fast Low Reasoning70.6% ±0.6$0.002931 s
55DeepSeek V3.264.8% ±2.0$0.00142 min
56Kimi K2 Instruct62.7% ±1.2$0.0132 min
57Qwen3 Maxqwen3-max-preview60.7% ±2.6$0.0233 min
58Grok 3 [Beta]58.8% ±1.4$0.09164 s
59Llama 3.3 Nemotron Super 49B V1 42e84561 Thinking53.5% ±1.6$03 min
60DeepSeek V3 032452.2% ±1.7$0.003251 s
61Gemini 2.5 Flash Preview Thinking51.5% ±1.9$0.00562 s
62Gemini 2.5 Flash (Preview)49.8% ±2.2$0.004558 s
63GPT-4.1 mini49.4% ±1.1$0.007933 s
64Claude 3.7 Sonnet Thinking44.6% ±1.7$0.3955 min
65Claude Opus 4.144.2% ±1.7$030 s
66Mistral Large 342.9% ±0.9$0.00612 min
67Mistral Medium42.3% ±1.3$066 s
68Gemini 2.5 Flash Lite Preview Thinking42.1% ±1.6$0.0005751 s
69Claude Opus 441.3% ±1.2$0.1137 s
70GPT-4.139.6% ±1.4$0.053 min
71Claude Sonnet 438.5% ±1.0$0.02223 s
72Grok 4 Fast Non Reasoning33.3% ±1.1$0.000815.9 s
73Gemini 2.0 Flash 00129.8% ±0.6$0.00111 s
74DeepSeek V327.5% ±0.7$0.003459 s
75Grok 4.1 Fast Non Reasoning27.1% ±1.9$0.0007917 s
76GPT-4.1 nano26.5% ±0.8$0.001212 s
77Gemini 2.5 Flash Lite Preview26.3% ±2.0$0.003219 s
78Llama4 Maverick Instruct Basic25.2% ±0.8$0.001116 s
79Claude 3.7 Sonnet22.3% ±1.4$0.02319 s
80Llama 4 Scout 17B 16E Instruct19.0% ±0.7$0.0008122 s
81Gemini 1.5 Pro 00218.8% ±0.5$0.004211 s
82Gemini 1.5 Flash 00217.3% ±0.9$0.000275.7 s
83Llama 3.3 70B Instruct Turbo16.0% ±1.1$0.001111 s
84Grok 215.2% ±0.8$0.01558 s
85Command A13.3% ±0.9$0.01223 s
86GPT-4o11.9% ±1.0$0.01216 s
87GPT-4o mini11.5% ±1.0$0.00129 s
88Claude 3.5 Sonnet10.0% ±0.9$0.00889.2 s
89Llama 3.3 Nemotron Super 49B V1 42e845619.4% ±0.5$016 s
90Mistral Large 24119.2% ±1.1$0.006320 s
91Mistral Small 24025.6% ±0.6$0.0007813 s
92Mistral Small 3.13.5% ±0.2$0.0002912 s
93Claude 3.5 Haiku3.3% ±0.3$0.00249.1 s
94Jamba Large 1.60.4% ±0.4$0.007719 s
94Jamba Mini 1.60.4% ±0.3$0.000466.6 s