TestsBenchmarks · Knowledge
Humanity's Last Exam
Accuracy on ~2,500 expert-written questions at the frontier of human knowledge (no tools unless noted).
% solvedHigher is betterCounts toward:Weights: Research & analysis ×0.5The test's websiteOfficial page ↗
Measured by independent testersIndependent Reported by the makerVendor-reported
| Thinking levelSetting | |||||||
|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | Anthropic | 61.4% | MaxClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 2 | Claude Fable 5.1 | Anthropic | 59.1% | MaxClaude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 3 | Muse Spark | Meta (Meta Superintelligence Labs, Muse) | 58.0% | MaxContemplating mode | Maker's own figureVendor-reportedMeta ↗ | — | 8 Apr 2026 |
| 4 | Claude Fable 5 | Anthropic | 57.4% | Extra highadaptive thinking, effort=xhigh | Maker's own figureVendor-reportedAnthropic ↗ | — | 1 Sep 2026 |
| 5 | Gemini 4 Argon | Google (Gemini / DeepMind) | 57.1% | HighGemini 4 Argon (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 6 | Claude Sonnet 5.5 | Anthropic | 55.0% | MaxClaude Sonnet 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 7 | Claude Opus 5 | Anthropic | 54.9% | MaxClaude Opus 5 (Adaptive Reasoning, Max Effort) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 8 | GPT-6 Astra | OpenAI | 54.8% | DefaultGPT 6 Astra | Independent testIndependentScale AI SEAL ↗ | — | 1 Oct 2026 |
| 9 | GPT-6.1 Sol | OpenAI | 52.9% | MaxGPT-6.1 Sol (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 10 | GPT-5.6 Sol | OpenAI | 49.5% | MaxGPT-5.6 Sol (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 11 | MiMo-V2.6-Pro | Xiaomi MiMo | 49.4% | DefaultMiMo-V2.6-Pro | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 12 | Claude Opus 4.8 | Anthropic | 48.7% | MaxClaude Opus 4.8 (Adaptive Reasoning, Max Effort) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 13 | Muse Spark 1.3 | Meta (Meta Superintelligence Labs, Muse) | 48.7% | MaxMuse Spark 1.3 (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 14 | Gemini 3.1 Pro (Preview) | Google (Gemini / DeepMind) | 48.4% | MaxDeep Think | Maker's own figureVendor-reportedGoogle ↗ | — | 12 Feb 2026 |
| 15 | Gemini 3.7 Flash | Google (Gemini / DeepMind) | 47.9% | HighGemini 3.7 Flash (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 16 | GPT-6 Sol | OpenAI | 47.9% | MaxGPT-6 Sol (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 17 | Gemini 3.8 Flash | Google (Gemini / DeepMind) | 47.8% | HighGemini 3.8 Flash (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 18 | Kimi K3 | Moonshot AI (Kimi) | 46.9% | MaxKimi K3 (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 19 | Step 5 Preview | StepFun | 46.5% | DefaultStep 5 Preview | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 20 | Muse Spark 1.1 | Meta (Meta Superintelligence Labs, Muse) | 46.2% | Extra highMuse Spark 1.1 (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 21 | GPT-5.5 | OpenAI | 45.8% | Extra highGPT-5.5 (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 22 | Muse Spark 1.2 | Meta (Meta Superintelligence Labs, Muse) | 45.5% | Extra highMuse Spark 1.2 (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 23 | GPT-5.4 Pro | OpenAI | 44.3% | Defaultgpt-5.4-pro-2026-03-05 | Independent testIndependentScale AI SEAL ↗ | — | 1 Oct 2026 |
| 24 | Grok 4.6 | SpaceXAI (formerly xAI) | 44.1% | Extra highGrok 4.6 (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 25 | GPT-5.4 | OpenAI | 43.7% | Extra highGPT-5.4 (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 26 | Qwen3.8 Max (0803) | Qwen (Alibaba) | 43.6% | Defaultthinking (default reasoning_effort=xhigh; eval setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | — | 3 Aug 2026 |
| 27 | Grok 4.7 | SpaceXAI (formerly xAI) | 43.1% | Extra highGrok 4.7 (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 28 | Qwen3.8 Max (0902) | Qwen (Alibaba) | 43.1% | DefaultQwen3.8 Max (0902) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 29 | GPT-5.5 Pro | OpenAI | 43.1% | Extra highGPT-5.5 Pro, reasoning effort=xhigh | Maker's own figureVendor-reportedOpenAI ↗ | — | 23 Apr 2026 |
| 30 | GPT-5.6 Terra | OpenAI | 42.9% | MaxGPT-5.6 Terra (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 31 | Grok 4.5 | SpaceXAI (formerly xAI) | 42.7% | HighGrok 4.5 (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 32 | Qwen3.8 2.4T A95B (open weights) | Qwen (Alibaba) | 42.4% | DefaultQwen3.8 2.4T A95B | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 33 | Claude Opus 4.7 | Anthropic | 42.3% | MaxClaude Opus 4.7 (Adaptive Reasoning, Max Effort) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 34 | GLM-5.3 | Z.ai (Zhipu) | 42.3% | MaxGLM-5.3 (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 35 | Qwen3.7 Max | Qwen (Alibaba) | 41.4% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | — | 16 May 2026 |
| 36 | Claude Sonnet 5 | Anthropic | 41.3% | MaxClaude Sonnet 5 (Adaptive Reasoning, Max Effort) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 37 | Gemini 3.5 Flash | Google (Gemini / DeepMind) | 41.3% | MediumGemini 3.5 Flash (Medium) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 38 | GLM-5.2 | Z.ai (Zhipu) | 41.1% | MaxGLM-5.2 (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 39 | DeepSeek V4 Pro (0813) | DeepSeek | 41.0% | MaxDeepSeek V4 Pro 0813 (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 40 | Gemini 3.6 Flash | Google (Gemini / DeepMind) | 40.8% | HighGemini 3.6 Flash (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 41 | GLM-5.3-Flash | Z.ai (Zhipu) | 39.9% | DefaultGLM 5.3 Flash | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 42 | GPT-5.6 Luna | OpenAI | 39.5% | MaxGPT-5.6 Luna (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 43 | DeepSeek V4.1 Flash | DeepSeek | 39.2% | MaxDeepSeek V4.1 Flash (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 44 | MiniMax M3 | MiniMax | 39.0% | DefaultMiniMax-M3 | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 45 | DeepSeek V4 Flash (0731) | DeepSeek | 38.6% | MaxDeepSeek V4 Flash 0731 (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 46 | GPT-6 Luna | OpenAI | 38.5% | MaxGPT-6 Luna (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 47 | Qwen3.8-Flash-Next | Qwen (Alibaba) | 38.0% | DefaultQwen3.8-Flash-Next | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 48 | DeepSeek V4 Pro (Preview, 0423) | DeepSeek | 37.7% | MaxThink Max | Maker's own figureVendor-reportedDeepSeek ↗ | — | 24 Apr 2026 |
| 49 | Gemini 3 Pro | Google (Gemini / DeepMind) | 37.5% | Defaultgemini-3-pro-preview | Independent testIndependentScale AI SEAL ↗ | — | 1 Oct 2026 |
| 50 | Qwen3.8 Flash | Qwen (Alibaba) | 35.9% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | — | 26 Aug 2026 |
| 51 | MiMo-V2.6-Flash | Xiaomi MiMo | 35.1% | DefaultMiMo-V2.6-Flash | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 52 | DeepSeek V4 Flash (Preview, 0423) | DeepSeek | 34.8% | MaxThink Max | Maker's own figureVendor-reportedDeepSeek ↗ | — | 24 Apr 2026 |
| 53 | Qwen3.7 Plus | Qwen (Alibaba) | 34.7% | Defaultthinking | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | — | 21 May 2026 |
| 54 | Kimi K2.6 | Moonshot AI (Kimi) | 34.7% | Defaultthinking | Maker's own figureVendor-reportedMoonshot AI ↗ | — | 20 Apr 2026 |
| 55 | Claude Sonnet 4.6 | Anthropic | 34.6% | Maxadaptive thinking, effort=max | Maker's own figureVendor-reportedAnthropic ↗ | — | 30 Jun 2026 |
| 56 | DeepSeek V4 Flash Vision Exp | DeepSeek | 34.5% | MaxDeepSeek V4 Flash Vision (Max) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 57 | Claude Opus 4.6 | Anthropic | 34.4% | Maxclaude-opus-4-6-thinking-max | Independent testIndependentScale AI SEAL ↗ | — | 1 Oct 2026 |
| 58 | Qwen3.8 27B | Qwen (Alibaba) | 33.9% | Extra highQwen3.8 27B (Xhigh) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 59 | Hy3 | Tencent Hunyuan | 33.5% | DefaultHy3 | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 60 | GLM-5.1 | Z.ai (Zhipu) | 31.0% | Defaultthinking | Maker's own figureVendor-reportedZ.ai ↗ | — | 7 Apr 2026 |
| 61 | Qwen3.6 Plus | Qwen (Alibaba) | 28.8% | Defaultthinking | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | — | 2 Apr 2026 |
| 62 | Nemotron 3 Ultra | NVIDIA | 28.4% | DefaultNemotron 3 Ultra 550B A55B (Reasoning) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 63 | Claude Opus 4.5 | Anthropic | 25.2% | Defaultclaude-opus-4-5-20251101-thinking | Independent testIndependentScale AI SEAL ↗ | — | 1 Oct 2026 |
| 64 | Kimi K2.5 | Moonshot AI (Kimi) | 24.4% | Defaultkimi-k2.5 | Independent testIndependentScale AI SEAL ↗ | — | 1 Oct 2026 |
| 65 | Qwen3.6 27B | Qwen (Alibaba) | 24.0% | Defaultthinking | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | — | 22 Apr 2026 |
| 66 | Gemma 4 31B IT | Google (Gemini / DeepMind) | 23.6% | DefaultGemma 4 31B (Reasoning) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 67 | Qwen3.6 35B A3B | Qwen (Alibaba) | 22.2% | DefaultQwen3.6 35B A3B (Reasoning) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 68 | Muse Glimmer 30B | Meta (Meta Superintelligence Labs, Muse) | 22.0% | HighMuse Glimmer (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 69 | Nemotron 3 Super 120B A12B | NVIDIA | 20.8% | DefaultNemotron 3 Super 120B A12B (Reasoning) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 70 | gpt-oss-120b | OpenAI | 19.6% | Highgpt-oss-120b (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 71 | Gemini 3.1 Flash-Lite | Google (Gemini / DeepMind) | 16.0% | High | Maker's own figureVendor-reportedGoogle ↗ | — | 3 Mar 2026 |
| 72 | Mistral Medium 3.5 | Mistral AI | 13.8% | DefaultMistral Medium 3.5 | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 73 | Command A+ | Cohere | 12.0% | DefaultCommand A+ | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 74 | gpt-oss-20b | OpenAI | 11.0% | Highgpt-oss-20b (High) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 75 | Nemotron 3.5 Lightning 30B-A3B | NVIDIA | 10.6% | DefaultNemotron 3.5 Lightning | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 76 | Mistral Small 4 | Mistral AI | 9.9% | HighMistral Small 4 (Reasoning) | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 77 | Granite 4.2 8B | IBM Granite | 9.7% | DefaultGranite 4.2 8B | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |
| 78 | Gemma 4 26B A4B IT | Google (Gemini / DeepMind) | 8.7% | DefaultThinking | Maker's own figureVendor-reportedGoogle ↗ | — | 2 Apr 2026 |
| 79 | Ministral 3 14B | Mistral AI | 4.6% | DefaultMinistral 3 14B | Independent testIndependentArtificial Analysis ↗ | — | 1 Oct 2026 |