AI models

OTIS Mock AIME: AI model scores

182 AI models on OTIS Mock AIME, per Epoch AI. It is a saturated benchmark: several models from different labs have already hit the ceiling.

OTIS Mock AIME, on the Epoch AI leaderboard (CC BY 4.0, latest version we checked), is a set of math problems in the style of the US AIME olympiad, in a practice version built so the model cannot lean on answers it may already have seen. The score is the share of correct answers. Here are the 182 models in this section with a score on OTIS Mock AIME, per Epoch AI. 16 models score 100%, so the table shows the scores without a rank.

OTIS Mock AIME

100%

Claude Fable 5

models

182

01

Model%Source
Claude Fable 5Anthropic · closed100% top scoreEpoch AI
Claude Fable 5.1Anthropic · closed100% top scoreEpoch AI
Claude Opus 5.5Anthropic · closed100% top scoreEpoch AI
Claude Sonnet 5.5Anthropic · closed100% top scoreEpoch AI
Gemini 4 ArgonGoogle · closed100% top scoreEpoch AI
Gemini 4 CarbonGoogle · closed100% top scoreEpoch AI
GPT-5.5OpenAI · closed100% top scoreEpoch AI
GPT-5.5 ProOpenAI · closed100% top scoreEpoch AI
GPT-5.6 SolOpenAI · closed100% top scoreEpoch AI
GPT-6 AstraOpenAI · closed100% top scoreEpoch AI
GPT-6 SolOpenAI · closed100% top scoreEpoch AI
GPT-6.1 SolOpenAI · closed100% top scoreEpoch AI
GPT-6.1 TerraOpenAI · closed100% top scoreEpoch AI
GPT-6.2 AstraOpenAI · closed100% top scoreEpoch AI
Kimi K3.1 ProMoonshot AI · open100% top scoreEpoch AI
Qwen3.8 MaxAlibaba · closed100% top scoreEpoch AI
Claude Fable 5.2Anthropic · closed99.7% statistical tie with the top scoreEpoch AI
GPT-5.6 TerraOpenAI · closed99.7% statistical tie with the top scoreEpoch AI
Mistral Large 4Mistral AI · open99.4% statistical tie with the top scoreEpoch AI
Muse WonderMeta · closed99.4% statistical tie with the top scoreEpoch AI
Doubao Seed Evolving (Nov 2026)ByteDance · closed99.2% statistical tie with the top scoreEpoch AI
Grok 4.6xAI · closed99.2% statistical tie with the top scoreEpoch AI
Muse Spark 1.3Meta · closed99.2% statistical tie with the top scoreEpoch AI
Claude Haiku 5.5Anthropic · closed98.9% statistical tie with the top scoreEpoch AI
Claude Opus 5Anthropic · closed98.9% statistical tie with the top scoreEpoch AI
Gemini 3.8 FlashGoogle · closed98.9% statistical tie with the top scoreEpoch AI
GPT-6 LunaOpenAI · closed98.9% statistical tie with the top scoreEpoch AI
Claude Opus 4.8Anthropic · closed98.3% statistical tie with the top scoreEpoch AI
DeepSeek V4.1 FlashDeepSeek · open98.3% statistical tie with the top scoreEpoch AI
GPT-5.6 LunaOpenAI · closed98.3% statistical tie with the top scoreEpoch AI
Grok 4.7xAI · closed98.1% statistical tie with the top scoreEpoch AI
Muse Spark OpenMeta · open98.1% statistical tie with the top scoreEpoch AI
Claude Opus 4.7Anthropic · closed97.8% statistical tie with the top scoreEpoch AI
GPT-5.4OpenAI · closed97.8% statistical tie with the top scoreEpoch AI
Grok 4.5xAI · closed97.8% statistical tie with the top scoreEpoch AI
Gemini 3.7 FlashGoogle · closed97.2% statistical tie with the top scoreEpoch AI
Kimi K3Moonshot AI · open97.2% statistical tie with the top scoreEpoch AI
DeepSeek V4 ProDeepSeek · open96.7% statistical tie with the top scoreEpoch AI
Doubao Seed 2.1 TurboByteDance · closed96.1% statistical tie with the top scoreEpoch AI
GPT-5.2OpenAI · closed96.1% statistical tie with the top scoreEpoch AI
Kimi K2.6Moonshot AI · open96.1% statistical tie with the top scoreEpoch AI
Doubao Seed 2.1 ProByteDance · closed95.6% statistical tie with the top scoreEpoch AI
Gemini 3 FlashGoogle · closed95.6% statistical tie with the top scoreEpoch AI
Gemini 3.1 ProGoogle · closed95.6% statistical tie with the top scoreEpoch AI
Gemini 3.5 FlashGoogle · closed95.6% statistical tie with the top scoreEpoch AI
Kimi K2.7 CodeMoonshot AI · open95.6% statistical tie with the top scoreEpoch AI
Qwen3.7 MaxAlibaba · closed95.6% statistical tie with the top scoreEpoch AI
Claude Sonnet 5Anthropic · closed94.7% statistical tie with the top scoreEpoch AI
Claude Opus 4.6Anthropic · closed94.4% statistical tie with the top scoreEpoch AI
DeepSeek V4 Flash 0731DeepSeek · open94.4% statistical tie with the top scoreEpoch AI
Gemini 3.6 FlashGoogle · closed94.2% statistical tie with the top scoreEpoch AI
Doubao-Seed-2.0 ProByteDance · closed93.9% statistical tie with the top scoreEpoch AI
GLM-5.3-FlashZhipu AI (Z.ai) · open93.9% statistical tie with the top scoreEpoch AI
GLM 5.1Zhipu AI (Z.ai) · open93.3% statistical tie with the top scoreEpoch AI
Grok 4.3 BetaxAI · closed93.3% statistical tie with the top scoreEpoch AI
Qwen3.6 PlusAlibaba · closed93.3% statistical tie with the top scoreEpoch AI
Qwen3.7 PlusAlibaba · closed93.3% statistical tie with the top scoreEpoch AI
Grok 4.20xAI · closed92.2% statistical tie with the top scoreEpoch AI
Kimi K2.5Moonshot AI · open92.2% statistical tie with the top scoreEpoch AI
Gemini 3 ProGoogle · closed91.4% statistical tie with the top scoreEpoch AI
GPT-5OpenAI · closed91.4% statistical tie with the top scoreEpoch AI
Doubao Seed 2.0 Lite (Apr 2026)ByteDance · closed91.1% statistical tie with the top scoreEpoch AI
GLM-5.3Zhipu AI (Z.ai) · open91.1% statistical tie with the top scoreEpoch AI
Qwen 3.6 Max (Preview)Alibaba · undisclosed91.1% statistical tie with the top scoreEpoch AI
Qwen3.6 27BAlibaba · open91.1% statistical tie with the top scoreEpoch AI
GPT-5.4 MiniOpenAI · closed88.9% statistical tie with the top scoreEpoch AI
gpt-oss-120bOpenAI · open88.9% statistical tie with the top scoreEpoch AI
Muse SparkMeta · closed88.9% statistical tie with the top scoreEpoch AI
Qwen3.5 397B A17BAlibaba · open88.9% statistical tie with the top scoreEpoch AI
GPT-5.1OpenAI · closed88.6% statistical tie with the top scoreEpoch AI
DeepSeek V3.2DeepSeek · open87.8% statistical tie with the top scoreEpoch AI
GPT-5.4 NanoOpenAI · closed87.8% statistical tie with the top scoreEpoch AI
GPT-5 MiniOpenAI · closed86.7% statistical tie with the top scoreEpoch AI
Nemotron 3 UltraNVIDIA · open86.7% statistical tie with the top scoreEpoch AI
Qwen3-235B-A22BAlibaba · open86.7% statistical tie with the top scoreEpoch AI
Qwen3.5 Plus 2026-02-15Alibaba · undisclosed86.7% statistical tie with the top scoreEpoch AI
Qwen3.6 35B A3BAlibaba · open86.7% statistical tie with the top scoreEpoch AI
Qwen3.7 FlashAlibaba · closed86.7% statistical tie with the top scoreEpoch AI
GLM-5.2Zhipu AI (Z.ai) · open86.4% statistical tie with the top scoreEpoch AI
Claude Opus 4.5Anthropic · closed86.1% statistical tie with the top scoreEpoch AI
Claude Sonnet 4.6Anthropic · closed85.8% statistical tie with the top scoreEpoch AI
Gemini 2.5 ProGoogle · closed84.7% statistical tie with the top scoreEpoch AI
o3OpenAI · closed84.4% statistical tie with the top scoreEpoch AI
Qwen3.5-FlashAlibaba · undisclosed84.4% statistical tie with the top scoreEpoch AI
Qwen3.6 FlashAlibaba · undisclosed84.4% statistical tie with the top scoreEpoch AI
Grok 4xAI · closed84% statistical tie with the top scoreEpoch AI
GLM 4.7Zhipu AI (Z.ai) · open83.3% statistical tie with the top scoreEpoch AI
Kimi K2 ThinkingMoonshot AI · open83.1% statistical tie with the top scoreEpoch AI
Gemma 4 26B A4BGoogle · open82.2% statistical tie with the top scoreEpoch AI
o4-miniOpenAI · closed81.7% statistical tie with the top scoreEpoch AI
GPT-5 NanoOpenAI · closed81.1% statistical tie with the top scoreEpoch AI
Gemini 3.1 Flash LiteGoogle · closed80% statistical tie with the top scoreEpoch AI
GLM 5Zhipu AI (Z.ai) · open80% statistical tie with the top scoreEpoch AI
Claude Sonnet 4.5Anthropic · closed77.8% statistical tie with the top scoreEpoch AI
Grok-3 minixAI · closed77.8% statistical tie with the top scoreEpoch AI
o3-miniOpenAI · closed76.9% statistical tie with the top scoreEpoch AI
o1OpenAI · closed73.3% statistical tie with the top scoreEpoch AI
Qwen3-MaxAlibaba · closed73.3% statistical tie with the top scoreEpoch AI
Claude Sonnet 4Anthropic · closed71.1% statistical tie with the top scoreEpoch AI
Gemini 3.5 Flash LiteGoogle · closed71.1% statistical tie with the top scoreEpoch AI
MiniMax M3MiniMax · open71.1% statistical tie with the top scoreEpoch AI
Qwen3.5-35B-A3BAlibaba · open70% statistical tie with the top scoreEpoch AI
Doubao-Seed-1.6ByteDance · closed69.4% statistical tie with the top scoreEpoch AI
Claude Opus 4.1Anthropic · closed68.9% Epoch AI
GPT-5.5 InstantOpenAI · closed68.1% Epoch AI
Qwen3 32BAlibaba · open66.9% Epoch AI
Claude Haiku 4.5Anthropic · closed66.7% Epoch AI
DeepSeek R1DeepSeek · open66.4% Epoch AI
Qwen3 14BAlibaba · open66.4% Epoch AI
gpt-oss-20bOpenAI · open65.3% Epoch AI
Claude Opus 4Anthropic · closed64.4% Epoch AI
Qwen3 30B A3BAlibaba · open62.8% Epoch AI
Qwen3.5-9BAlibaba · open61.7% Epoch AI
QwQ-32BAlibaba · open59.2% Epoch AI
Claude 3.7 SonnetAnthropic · closed57.8% Epoch AI
Gemini 2.0 Flash ThinkingGoogle · closed57.8% Epoch AI
Qwen3-8BAlibaba · open56.1% Epoch AI
Qwen3.5-4BAlibaba · open55.8% Epoch AI
DeepSeek-R1-Distill-Qwen-32BDeepSeek · open55.6% Epoch AI
Grok 3xAI · closed55.6% Epoch AI
DeepSeek-R1-Distill-Llama-70BDeepSeek · open51.4% Epoch AI
DeepSeek-R1-Distill-Qwen-14BDeepSeek · open50.6% Epoch AI
o1-miniOpenAI · closed46.9% Epoch AI
GPT-4.1 MiniOpenAI · closed44.7% Epoch AI
GPT-4.1OpenAI · closed38.3% Epoch AI
DeepSeek V3DeepSeek · open37.8% Epoch AI
GPT-4.5OpenAI · closed37.8% Epoch AI
Mistral Medium 3Mistral AI · closed32.2% Epoch AI
Gemini 2.0 FlashGoogle · closed31.1% Epoch AI
o1-previewOpenAI · closed31.1% Epoch AI
Mistral Small 3.2Mistral AI · open30.3% Epoch AI
Magistral Small 1.0Mistral AI · open30% Epoch AI
GPT-4.1 NanoOpenAI · closed28.9% Epoch AI
Magistral Small 1.2Mistral AI · open28.1% Epoch AI
Gemini 1.5 ProGoogle · closed23.1% Epoch AI
Gemma 3 27BGoogle · open22.5% Epoch AI
DeepSeek-R1-Distill-Qwen-1.5BDeepSeek · open21.4% Epoch AI
Llama 4 MaverickMeta · open20.6% Epoch AI
Qwen-PlusAlibaba · closed17.8% Epoch AI
Gemma 3 12BGoogle · open16.7% Epoch AI
Gemini 1.5 FlashGoogle · closed16.2% Epoch AI
Qwen2.5-MaxAlibaba · closed16.1% Epoch AI
Phi 4Microsoft · open13.8% Epoch AI
Grok-2xAI · closed11.5% Epoch AI
Llama 3.1-405BMeta · open9.7% Epoch AI
Claude 3.5 Sonnet (October 2024)Anthropic · closed8.5% Epoch AI
Qwen2.5-72BAlibaba · open8.1% Epoch AI
Qwen3-1.7BAlibaba · open8.1% Epoch AI
Llama 4 ScoutMeta · open7.8% Epoch AI
Mistral Large 2Mistral AI · open7.8% Epoch AI
Gemma 3 4BGoogle · open7.5% Epoch AI
Qwen2.5-32BAlibaba · open7.4% Epoch AI
GPT-4o-miniOpenAI · closed6.9% Epoch AI
GPT-4 Turbo (Apr 2024)OpenAI · closed6.7% Epoch AI
Mistral Small 3Mistral AI · open6.7% Epoch AI
Claude 3.5 SonnetAnthropic · closed6.5% Epoch AI
Qwen-TurboAlibaba · closed6.1% Epoch AI
Mistral Small 3.1Mistral AI · open5.8% Epoch AI
Llama 3.3 70BMeta · open5.1% Epoch AI
Claude 3 OpusAnthropic · closed4.7% Epoch AI
Gemini 1.5 Flash 8BGoogle · closed4.6% Epoch AI
Tulu 3 (Tülu 3) 70BAi2 (Allen Institute) · open4.4% Epoch AI
Claude 3.5 HaikuAnthropic · closed4.3% Epoch AI
Llama 3-70BMeta · open4.3% Epoch AI
Llama 3.1-70BMeta · open3.6% Epoch AI
Llama 3.2 90BMeta · open2.6% Epoch AI
Claude 2Anthropic · closed2.5% Epoch AI
Claude 3 SonnetAnthropic · closed2.5% Epoch AI
Qwen2.5-7BAlibaba · open2.5% Epoch AI
Claude 2.1Anthropic · closed1.9% Epoch AI
Llama 3-8BMeta · open1.9% Epoch AI
Mistral LargeMistral AI · closed1.9% Epoch AI
Claude 3 HaikuAnthropic · closed1.8% Epoch AI
Llama 3.1-8BMeta · open1.7% Epoch AI
Gemma 2 27BGoogle · open1.4% Epoch AI
Gemini 1.0 ProGoogle · closed1.1% Epoch AI
Gemma 3 1BGoogle · open1.1% Epoch AI
DeepSeek LLM 67BDeepSeek · open0.8% Epoch AI
Gemma 2 9BGoogle · open0.6% Epoch AI
Llama 3.2 1BMeta · open0.6% Epoch AI
Mistral 7BMistral AI · open0.3% Epoch AI
Llama 2-70BMeta · open0% Epoch AI

FAQ

Which model has the top score on OTIS Mock AIME?

16 models score 100% on OTIS Mock AIME, per Epoch AI. It is a saturated benchmark: several models from different labs have already hit the ceiling, and the table shows the scores without a rank.

What does OTIS Mock AIME measure?

OTIS Mock AIME, on the Epoch AI leaderboard (CC BY 4.0, latest version we checked), is a set of math problems in the style of the US AIME olympiad, in a practice version built so the model cannot lean on answers it may already have seen. The score is the share of correct answers.

Which models are in this ranking?

The 182 models in this section with a published score on OTIS Mock AIME, per Epoch AI. No version inherits another's score.

How to read

  1. Benchmark scores are percentages, and higher is better.
  2. Rank is shown as a range, «1st of 182». Within the range, the gap to neighboring models falls within the benchmark's margin. The «top score» badge marks the model with the benchmark's top score, and the «statistical tie with the top score» badge marks models whose gap to it is within the margin.
  3. On a saturated benchmark, the table shows the scores with the badge and without a rank order.

Sources

  • Epoch AI, CC BY 4.0, latest version we checked.
  • OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
  • OpenRouter, dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.

How to cite

If you use this data, cite the page and the date you accessed it.

Adriano Meirinho. "OTIS Mock AIME: AI model scores". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/benchmark/otis-aime/. Accessed November 21, 2026.

Third-party data follows the license of each source. The full table is available as CSV and JSON.

Markdown