OpenAI GPT-4.1 (high)72.4% vals.ai for GPT-4.1 (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-4.1 on MMMU-Pro Vals AI harness 10002 OpenAI OpenAI o3 (high)80.4% vals.ai for OpenAI o3 (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for OpenAI o3 on MMMU-Pro Vals AI harness 10003 OpenAI OpenAI o4-mini (high)79.7% vals.ai for OpenAI o4-mini (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for OpenAI o4-mini on MMMU-Pro Vals AI harness 10004 OpenAI GPT-5.4 (xhigh)87.5% vals.ai for GPT-5.4 (xhigh) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.4 on MMMU-Pro Vals AI harness 10007 OpenAI GPT-5.5 (xhigh)88.3% vals.ai for GPT-5.5 (xhigh) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.5 on MMMU-Pro Vals AI harness 10008 OpenAI GPT-5.6 Luna (max)85.0% vals.ai for GPT-5.6 Luna (max) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.6 Luna on MMMU-Pro Vals AI harness 10009 OpenAI GPT-5.6 Sol (max)88.8% vals.ai for GPT-5.6 Sol (max) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.6 Sol on MMMU-Pro Vals AI harness 10010 OpenAI GPT-5.6 Terra (max)86.5% vals.ai for GPT-5.6 Terra (max) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.6 Terra on MMMU-Pro Vals AI harness 10011 OpenAI GPT-5.2 (xhigh)86.7% vals.ai for GPT-5.2 (xhigh) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.2 on MMMU-Pro Vals AI harness 10012 OpenAI GPT-4.1 mini (high)70.5% vals.ai for GPT-4.1 mini (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-4.1 mini on MMMU-Pro Vals AI harness 10016 OpenAI GPT-4.1 nano (high)55.1% vals.ai for GPT-4.1 nano (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-4.1 nano on MMMU-Pro Vals AI harness 10017 OpenAI GPT-5 (high)81.5% vals.ai for GPT-5 (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5 on MMMU-Pro Vals AI harness 10018 OpenAI GPT-5 mini (high)78.9% vals.ai for GPT-5 mini (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5 mini on MMMU-Pro Vals AI harness 10019 OpenAI GPT-5 nano (high)70.9% vals.ai for GPT-5 nano (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5 nano on MMMU-Pro Vals AI harness 10020 OpenAI GPT-5.4 mini (xhigh)79.2% vals.ai for GPT-5.4 mini (xhigh) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.4 mini on MMMU-Pro Vals AI harness 10021 OpenAI GPT-5.4 nano (high)73.6% vals.ai for GPT-5.4 nano (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GPT-5.4 nano on MMMU-Pro Vals AI harness 10022 Moonshot AI Kimi K2.5 (thinking)84.3% vals.ai for Kimi K2.5 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Kimi K2.5 on MMMU-Pro Vals AI harness 120001 Moonshot AI Kimi K2.6 86.3% vals.ai for Kimi K2.6 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Kimi K2.6 on MMMU-Pro Vals AI harness 120002 Moonshot AI Kimi K3 88.2% vals.ai for Kimi K3 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Kimi K3 on MMMU-Pro Vals AI harness 120003 Alibaba (Tongyi) Qwen3.6-Plus 84.2% vals.ai for Qwen3.6-Plus on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Qwen3.6-Plus on MMMU-Pro Vals AI harness 130004 Alibaba (Tongyi) Qwen3.8-Max 88.0% vals.ai for Qwen3.8-Max on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Qwen3.8-Max on MMMU-Pro Vals AI harness 130008 MiniMax MiniMax M3 81.2% vals.ai for MiniMax M3 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for MiniMax M3 on MMMU-Pro Vals AI harness 140003 Z.ai GLM-5.3-Flash (max)86.0% vals.ai for GLM-5.3-Flash (max) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for GLM-5.3-Flash on MMMU-Pro Vals AI harness 150006 Thinking Machines Lab Inkling (0.99)78.5% vals.ai for Inkling (0.99) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Inkling on MMMU-Pro Vals AI harness 160001 Anthropic Claude 3.7 Sonnet 71.5% vals.ai for Claude 3.7 Sonnet on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude 3.7 Sonnet on MMMU-Pro Vals AI harness 20001 Anthropic Claude 3.7 Sonnet (thinking)75.1% vals.ai for Claude 3.7 Sonnet (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude 3.7 Sonnet on MMMU-Pro Vals AI harness 20001 Anthropic Claude Opus 4 73.3% vals.ai for Claude Opus 4 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4 on MMMU-Pro Vals AI harness 20002 Anthropic Claude Sonnet 4 (thinking)74.9% vals.ai for Claude Sonnet 4 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Sonnet 4 on MMMU-Pro Vals AI harness 20003 Anthropic Claude Sonnet 4 72.4% vals.ai for Claude Sonnet 4 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Sonnet 4 on MMMU-Pro Vals AI harness 20003 Anthropic Claude Fable 5.1 90.6% vals.ai for Claude Fable 5.1 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Fable 5.1 on MMMU-Pro Vals AI harness 20004 Anthropic Claude Sonnet 4.5 (thinking)79.3% vals.ai for Claude Sonnet 4.5 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Sonnet 4.5 on MMMU-Pro Vals AI harness 20005 Anthropic Claude Haiku 4.5 (thinking)46.1% vals.ai for Claude Haiku 4.5 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Haiku 4.5 on MMMU-Pro Vals AI harness 20006 Anthropic Claude Opus 4.5 81.1% vals.ai for Claude Opus 4.5 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.5 on MMMU-Pro Vals AI harness 20007 Anthropic Claude Opus 4.5 (thinking)82.9% vals.ai for Claude Opus 4.5 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.5 on MMMU-Pro Vals AI harness 20007 Anthropic Claude Opus 4.6 (thinking)83.9% vals.ai for Claude Opus 4.6 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.6 on MMMU-Pro Vals AI harness 20008 Anthropic Claude Sonnet 4.6 83.6% vals.ai for Claude Sonnet 4.6 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Sonnet 4.6 on MMMU-Pro Vals AI harness 20009 Anthropic Claude Opus 4.7 85.5% vals.ai for Claude Opus 4.7 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.7 on MMMU-Pro Vals AI harness 20010 Anthropic Claude Opus 4.8 86.6% vals.ai for Claude Opus 4.8 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.8 on MMMU-Pro Vals AI harness 20011 Anthropic Claude Fable 5 89.3% vals.ai for Claude Fable 5 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Fable 5 on MMMU-Pro Vals AI harness 20012 Anthropic Claude Sonnet 5 83.0% vals.ai for Claude Sonnet 5 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Sonnet 5 on MMMU-Pro Vals AI harness 20013 Anthropic Claude Opus 5 89.9% vals.ai for Claude Opus 5 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 5 on MMMU-Pro Vals AI harness 20014 Anthropic Claude Opus 4.1 (thinking)77.5% vals.ai for Claude Opus 4.1 (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.1 on MMMU-Pro Vals AI harness 20017 Anthropic Claude Opus 4.1 73.7% vals.ai for Claude Opus 4.1 on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Claude Opus 4.1 on MMMU-Pro Vals AI harness 20017 Google (DeepMind) Gemini 2.0 Flash 69.8% vals.ai for Gemini 2.0 Flash on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 2.0 Flash on MMMU-Pro Vals AI harness 30001 Google (DeepMind) Gemini 2.5 Pro 81.3% vals.ai for Gemini 2.5 Pro on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 2.5 Pro on MMMU-Pro Vals AI harness 30002 Google (DeepMind) Gemini 2.5 Flash 68.0% vals.ai for Gemini 2.5 Flash on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 2.5 Flash on MMMU-Pro Vals AI harness 30003 Google (DeepMind) Gemini 2.5 Flash (thinking)71.9% vals.ai for Gemini 2.5 Flash (thinking) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 2.5 Flash on MMMU-Pro Vals AI harness 30003 Google (DeepMind) Gemini 3.8 Flash (high)89.1% vals.ai for Gemini 3.8 Flash (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 3.8 Flash on MMMU-Pro Vals AI harness 30004 Google (DeepMind) Gemini 3.1 Pro (high)88.2% vals.ai for Gemini 3.1 Pro (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 3.1 Pro on MMMU-Pro Vals AI harness 30005 Google (DeepMind) Gemini 3.5 Flash (high)88.3% vals.ai for Gemini 3.5 Flash (high) on MMMU-Pro Vals AI harness (opens in a new tab) Archived copy (opens in a new tab) MMMU-Pro accuracy Setting: Standard (4 options) Reported September 1, 2026 Benchmark leaderboard Vals AI Independent Record for Gemini 3.5 Flash on MMMU-Pro Vals AI harness 30006