evals.report
BenchmarksLabsCompareRun guidesIn the wild
BenchmarksReasoning

AIME (OTIS Mock)

Competition mathematics in the AIME format (Epoch AI's OTIS Mock AIME 2024-2025 set), a high-signal short-answer math reasoning benchmark.

ReasoningaccuracyHigher is better
ScoresAboutRun this benchmark

What is AIME (OTIS Mock)?

Competition mathematics in the AIME format (Epoch AI's OTIS Mock AIME 2024-2025 set), a high-signal short-answer math reasoning benchmark. evals.report tracks reported AIME (OTIS Mock) scores with the model, source, status, date, and run caveats attached — official leaderboard scores, vendor-reported launches, and clearly labeled community runs.

Top reported AIME (OTIS Mock) score: GPT-5.5 Pro — 100.0% (accuracy).

ModelLabScore↓Source modelStatusDate
GPT-5.5 ProOpenAI100.0%GPT-5.5 ProOfficialApr 23, 2026Details
GPT-5.5OpenAI100.0%GPT-5.5OfficialApr 23, 2026Details
Claude Opus 4.7Anthropic97.8%Claude Opus 4.7OfficialApr 16, 2026Details
Kimi K2.6OpenMoonshot AI96.1%Kimi K2.6OfficialApr 20, 2026Details
GPT-5.2OpenAI96.1%GPT-5.2OfficialDec 11, 2025Details
Gemini 3.1 Pro PreviewGoogle DeepMind95.6%Gemini 3.1 ProOfficialFeb 19, 2026Details
Gemini 3.5 FlashGoogle DeepMind95.6%Gemini 3.5 FlashOfficialMay 19, 2026Details
GPT-5.4OpenAI95.3%GPT-5.4OfficialMar 5, 2026Details
Claude Opus 4.6Anthropic94.4%Claude Opus 4.6OfficialFeb 5, 2026Details
Gemini 3 FlashGoogle DeepMind92.8%Gemini 3 FlashOfficialDec 17, 2025Details
GLM-5.1OpenZ.ai92.2%GLM-5.1OfficialApr 7, 2026Details
Kimi K2.5OpenMoonshot AI92.2%Kimi K2.5OfficialJan 27, 2026Details
Gemini 3 ProGoogle DeepMind91.4%Gemini 3 ProOfficialNov 18, 2025Details
GPT-5OpenAI91.4%GPT-5OfficialAug 7, 2025Details
Qwen 3.6 Max PreviewAlibaba / Qwen91.1%Qwen 3.6 Max (Preview)OfficialApr 20, 2026Details
Qwen 3.6 PlusAlibaba / Qwen90.6%Qwen 3.6 PlusOfficialApr 2, 2026Details
Muse SparkMeta88.9%Muse SparkOfficialApr 8, 2026Details
GPT-OSS-120BOpenOpenAI88.9%gpt-oss-120bOfficialAug 5, 2025Details
GPT-5.1OpenAI88.6%GPT-5.1OfficialNov 12, 2025Details
DeepSeek V3.2OpenDeepSeek87.8%DeepSeek-V3.2OfficialDec 1, 2025Details
Qwen3 235B A22B Instruct 2507OpenAlibaba / Qwen86.7%Qwen3-235B-A22B (Jul 2025)OfficialJul 21, 2025Details
GPT-5 miniOpenAI86.7%GPT-5 miniOfficialAug 7, 2025Details
Claude Opus 4.5Anthropic86.1%Claude Opus 4.5OfficialNov 24, 2025Details
Claude Sonnet 4.6Anthropic85.8%Claude Sonnet 4.6OfficialFeb 17, 2026Details
Gemini 2.5 ProGoogle DeepMind84.2%Gemini 2.5 Pro (Jun 2025)OfficialMar 25, 2025Details
Grok 4xAI84.0%Grok 4OfficialJul 9, 2025Details
o3OpenAI83.9%o3OfficialApr 16, 2025Details
o4-miniOpenAI81.7%o4-miniOfficialApr 16, 2025Details
GLM-5OpenZ.ai80.0%GLM-5OfficialFeb 11, 2026Details
Claude Sonnet 4.5Anthropic77.8%Claude Sonnet 4.5OfficialSep 29, 2025Details
Qwen3 MaxAlibaba / Qwen73.3%Qwen3-MaxOfficialSep 5, 2025Details
Gemini 2.5 FlashGoogle DeepMind73.1%Gemini 2.5 FlashOfficialApr 17, 2025Details
Claude Sonnet 4Anthropic71.1%Claude Sonnet 4OfficialMay 22, 2025Details
Claude Opus 4.1Anthropic68.9%Claude Opus 4.1OfficialAug 5, 2025Details
Claude Haiku 4.5Anthropic66.7%Claude Haiku 4.5OfficialOct 15, 2025Details
Claude Opus 4Anthropic64.4%Claude Opus 4OfficialMay 22, 2025Details
Claude 3.7 SonnetAnthropic57.8%Claude 3.7 SonnetOfficialFeb 24, 2025Details
DeepSeek R1OpenDeepSeek53.3%DeepSeek-R1OfficialJan 20, 2025Details
GPT-4.1OpenAI38.3%GPT-4.1OfficialApr 14, 2025Details
Gemini 2.0 FlashGoogle DeepMind31.1%Gemini 2.0 FlashOfficialDec 11, 2024Details
Gemini 1.5 ProGoogle DeepMind23.1%Gemini 1.5 ProOfficialFeb 15, 2024Details
Llama 4 MaverickOpenMeta20.6%Llama 4 MaverickOfficialApr 5, 2025Details
DeepSeek V3OpenDeepSeek15.8%DeepSeek-V3OfficialDec 26, 2024Details
Llama 3.1 405BOpenMeta9.7%Llama 3.1-405BOfficialJul 23, 2024Details
Llama 4 ScoutOpenMeta7.8%Llama 4 ScoutOfficialApr 5, 2025Details
Claude 3.5 SonnetAnthropic6.5%Claude 3.5 SonnetOfficialJun 20, 2024Details
GPT-4oOpenAI6.4%GPT-4oOfficialMay 13, 2024Details
Mistral LargeMistral AI1.9%Mistral LargeOfficialFeb 26, 2024Details

Each row reports the model’s accuracy on AIME (OTIS Mock). Click a row for the full run context.