벤치마크 상세
Humanity's Last Exam
고난도 학술 문제: frontier 모델의 한계, 어려운 전문 질문, 장기 reasoning 비교
공식 리더보드 보기
텍스트 2,158문항에 대한 AA의 평가입니다. 이미지 포함 전체 HLE나 도구 사용 결과와 구분합니다. 원문 모델명에 담긴 추론 강도와 fallback 설정을 그대로 보존합니다.
모든 공개 결과
모델 검색 · 점수순확인 2026. 09. 07.AA · 2,158 text-only questions · May 2025 revision% pass@1
602개 결과원문 ↗
| 순서 | 모델 / 실행 구성 | 점수 | 평가 조건·근거 |
|---|---|---|---|
| 1 | Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic | 59.13 | AA · 2,158 text-only questions · May 2025 revision |
| 2 | Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic | 58.71 | AA · 2,158 text-only questions · May 2025 revision |
| 3 | Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic | 55.93 | AA · 2,158 text-only questions · May 2025 revision |
| 4 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic | 55.47 | AA · 2,158 text-only questions · May 2025 revision |
| 5 | Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic | 54.87 | AA · 2,158 text-only questions · May 2025 revision |
| 6 | GPT-6 Astra (max)OpenAI | 54.68 | AA · 2,158 text-only questions · May 2025 revision |
| 7 | GPT-6 Astra (xhigh)OpenAI | 54.59 | AA · 2,158 text-only questions · May 2025 revision |
| 8 | Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic | 54.4 | AA · 2,158 text-only questions · May 2025 revision |
| 9 | Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic | 53.8 | AA · 2,158 text-only questions · May 2025 revision |
| 10 | GPT-6 Astra (high)OpenAI | 53.06 | AA · 2,158 text-only questions · May 2025 revision |
| 11 | Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic | 52.83 | AA · 2,158 text-only questions · May 2025 revision |
| 12 | GPT-6 Astra (medium)OpenAI | 52.73 | AA · 2,158 text-only questions · May 2025 revision |
| 13 | Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic | 51.3 | AA · 2,158 text-only questions · May 2025 revision |
| 14 | GPT-5.6 Sol (max)OpenAI | 49.49 | AA · 2,158 text-only questions · May 2025 revision |
| 15 | GPT-6 Astra (low)OpenAI | 49.21 | AA · 2,158 text-only questions · May 2025 revision |
| 16 | Muse Spark 1.3 (max)Meta | 49.07 | AA · 2,158 text-only questions · May 2025 revision |
| 17 | Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic | 48.89 | AA · 2,158 text-only questions · May 2025 revision |
| 18 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic | 48.66 | AA · 2,158 text-only questions · May 2025 revision |
| 19 | Gemini 3.7 Flash (high)Google | 47.87 | AA · 2,158 text-only questions · May 2025 revision |
| 20 | Gemini 3.8 Flash (high)Google | 47.82 | AA · 2,158 text-only questions · May 2025 revision |
| 21 | Muse Spark 1.3 (xhigh)Meta | 47.45 | AA · 2,158 text-only questions · May 2025 revision |
| 22 | GPT-5.6 Sol (xhigh)OpenAI | 47.31 | AA · 2,158 text-only questions · May 2025 revision |
| 23 | Gemini 3.1 Pro PreviewGoogle | 47.03 | AA · 2,158 text-only questions · May 2025 revision |
| 24 | Kimi K3 (max)Kimi | 46.9 | AA · 2,158 text-only questions · May 2025 revision |
| 25 | Muse Spark 1.1 (xhigh)Meta | 46.2 | AA · 2,158 text-only questions · May 2025 revision |
| 26 | GPT-5.6 Sol (high)OpenAI | 46.01 | AA · 2,158 text-only questions · May 2025 revision |
| 27 | GPT-5.5 (xhigh)OpenAI | 45.78 | AA · 2,158 text-only questions · May 2025 revision |
| 28 | Muse Spark 1.2 (xhigh)Meta | 45.46 | AA · 2,158 text-only questions · May 2025 revision |
| 29 | GPT-5.5 (high)OpenAI | 45.04 | AA · 2,158 text-only questions · May 2025 revision |
| 30 | Grok 4.6 (xhigh)SpaceXAI | 44.07 | AA · 2,158 text-only questions · May 2025 revision |
| 31 | GPT-5.4 (xhigh)OpenAI | 43.74 | AA · 2,158 text-only questions · May 2025 revision |
| 32 | Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic | 43.42 | AA · 2,158 text-only questions · May 2025 revision |
| 33 | Qwen3.8 MaxAlibaba | 43.05 | AA · 2,158 text-only questions · May 2025 revision |
| 34 | GPT-5.6 Terra (max)OpenAI | 42.91 | AA · 2,158 text-only questions · May 2025 revision |
| 35 | Grok 4.6 (high)SpaceXAI | 42.91 | AA · 2,158 text-only questions · May 2025 revision |
| 36 | Gemini 3.5 Flash (high)Google | 42.68 | AA · 2,158 text-only questions · May 2025 revision |
| 37 | Grok 4.5 (high)SpaceXAI | 42.68 | AA · 2,158 text-only questions · May 2025 revision |
| 38 | GPT-5.3 Codex (xhigh)OpenAI | 42.49 | AA · 2,158 text-only questions · May 2025 revision |
| 39 | Qwen3.8 2.4T A95BAlibaba | 42.45 | AA · 2,158 text-only questions · May 2025 revision |
| 40 | GPT-5.5 (medium)OpenAI | 42.35 | AA · 2,158 text-only questions · May 2025 revision |
| 41 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic | 42.31 | AA · 2,158 text-only questions · May 2025 revision |
| 42 | GLM-5.3 (max)Z AI | 42.26 | AA · 2,158 text-only questions · May 2025 revision |
| 43 | GPT-5.6 Sol (medium)OpenAI | 42.22 | AA · 2,158 text-only questions · May 2025 revision |
| 44 | Grok 4.6 (medium)SpaceXAI | 42.12 | AA · 2,158 text-only questions · May 2025 revision |
| 45 | Gemini 3.8 Flash (medium)Google | 42.12 | AA · 2,158 text-only questions · May 2025 revision |
| 46 | GPT-5.6 Terra (xhigh)OpenAI | 41.89 | AA · 2,158 text-only questions · May 2025 revision |
| 47 | Gemini 3.5 Flash (medium)Google | 41.33 | AA · 2,158 text-only questions · May 2025 revision |
| 48 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic | 41.29 | AA · 2,158 text-only questions · May 2025 revision |
| 49 | GLM-5.2 (max)Z AI | 41.15 | AA · 2,158 text-only questions · May 2025 revision |
| 50 | DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek | 41.01 | AA · 2,158 text-only questions · May 2025 revision |
| 51 | Gemini 3.6 Flash (high)Google | 40.82 | AA · 2,158 text-only questions · May 2025 revision |
| 52 | Muse SparkMeta | 40.69 | AA · 2,158 text-only questions · May 2025 revision |
| 53 | Qwen3.7 MaxAlibaba | 40.5 | AA · 2,158 text-only questions · May 2025 revision |
| 54 | Motif 3 (Beta)Motif Technologies | 40.36 | AA · 2,158 text-only questions · May 2025 revision |
| 55 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic | 39.94 | AA · 2,158 text-only questions · May 2025 revision |
| 56 | GLM-5.3-FlashZ AI | 39.85 | AA · 2,158 text-only questions · May 2025 revision |
| 57 | Gemini 3 Pro Preview (high)Google | 39.71 | AA · 2,158 text-only questions · May 2025 revision |
| 58 | GPT-5.6 Luna (max)OpenAI | 39.48 | AA · 2,158 text-only questions · May 2025 revision |
| 59 | GPT-5.6 Sol (low)OpenAI | 39.39 | AA · 2,158 text-only questions · May 2025 revision |
| 60 | MiniMax-M3MiniMax | 38.97 | AA · 2,158 text-only questions · May 2025 revision |
| 61 | Gemini 3.7 Flash (medium)Google | 38.97 | AA · 2,158 text-only questions · May 2025 revision |
| 62 | DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek | 38.55 | AA · 2,158 text-only questions · May 2025 revision |
| 63 | GPT-5.6 Terra (high)OpenAI | 38.51 | AA · 2,158 text-only questions · May 2025 revision |
| 64 | Grok Build 0.1 0616SpaceXAI | 38.28 | AA · 2,158 text-only questions · May 2025 revision |
| 65 | Qwen3.8-Flash-NextAlibaba | 38.04 | AA · 2,158 text-only questions · May 2025 revision |
| 66 | GPT-5.2 (xhigh)OpenAI | 37.67 | AA · 2,158 text-only questions · May 2025 revision |
| 67 | DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek | 37.53 | AA · 2,158 text-only questions · May 2025 revision |
| 68 | Agnes 2.5 Pro BetaSapiens AI | 37.53 | AA · 2,158 text-only questions · May 2025 revision |
| 69 | Kimi K2.6Kimi | 37.49 | AA · 2,158 text-only questions · May 2025 revision |
| 70 | Grok 4.3 (high)SpaceXAI | 37.21 | AA · 2,158 text-only questions · May 2025 revision |
| 71 | GPT-6 Astra (Non-reasoning)OpenAI | 37.12 | AA · 2,158 text-only questions · May 2025 revision |
| 72 | Gemini 3.8 Flash (low)Google | 37.07 | AA · 2,158 text-only questions · May 2025 revision |
| 73 | GPT-5.6 Luna (xhigh)OpenAI | 36.98 | AA · 2,158 text-only questions · May 2025 revision |
| 74 | Motif 3Motif Technologies | 36.98 | AA · 2,158 text-only questions · May 2025 revision |
| 75 | Gemini 3 Flash Preview (Reasoning)Google | 36.56 | AA · 2,158 text-only questions · May 2025 revision |
| 76 | GPT-5.2 Codex (xhigh)OpenAI | 35.73 | AA · 2,158 text-only questions · May 2025 revision |
| 77 | MiMo-V2.5-ProXiaomi | 35.68 | AA · 2,158 text-only questions · May 2025 revision |
| 78 | Qwen3.7 PlusAlibaba | 35.63 | AA · 2,158 text-only questions · May 2025 revision |
| 79 | DeepSeek V4 Pro (Reasoning, High Effort)DeepSeek | 35.22 | AA · 2,158 text-only questions · May 2025 revision |
| 80 | Gemini 3.7 Flash (low)Google | 35.13 | AA · 2,158 text-only questions · May 2025 revision |
| 81 | Kimi K2.7 CodeKimi | 35.03 | AA · 2,158 text-only questions · May 2025 revision |
| 82 | DeepSeek V4 Flash (Reasoning, Max Effort)DeepSeek | 34.85 | AA · 2,158 text-only questions · May 2025 revision |
| 83 | Grok 4.20 0309 v2 (Reasoning)SpaceXAI | 34.52 | AA · 2,158 text-only questions · May 2025 revision |
| 84 | DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek | 34.48 | AA · 2,158 text-only questions · May 2025 revision |
| 85 | Apodex 1.1Apodex | 34.06 | AA · 2,158 text-only questions · May 2025 revision |
| 86 | Qwen3.8 27B (xhigh)Alibaba | 33.92 | AA · 2,158 text-only questions · May 2025 revision |
| 87 | Nex-N2-ProNex AGI | 33.73 | AA · 2,158 text-only questions · May 2025 revision |
| 88 | LongCat 2.0LongCat | 33.69 | AA · 2,158 text-only questions · May 2025 revision |
| 89 | KAT-Coder-Pro V1KwaiKAT | 33.64 | AA · 2,158 text-only questions · May 2025 revision |
| 90 | Agnes 2.5 Pro AlphaSapiens AI | 33.6 | AA · 2,158 text-only questions · May 2025 revision |
| 91 | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic | 33.6 | AA · 2,158 text-only questions · May 2025 revision |
| 92 | Hy3Tencent | 33.46 | AA · 2,158 text-only questions · May 2025 revision |
| 93 | GPT-5.6 Luna (high)OpenAI | 33.41 | AA · 2,158 text-only questions · May 2025 revision |
| 94 | Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic | 33.32 | AA · 2,158 text-only questions · May 2025 revision |
| 95 | Inkling SmallThinking Machines | 33.32 | AA · 2,158 text-only questions · May 2025 revision |
| 96 | GPT-5.6 Terra (medium)OpenAI | 33.27 | AA · 2,158 text-only questions · May 2025 revision |
| 97 | GPT-5.5 (low)OpenAI | 32.67 | AA · 2,158 text-only questions · May 2025 revision |
| 98 | Grok 4.20 0309 (Reasoning)SpaceXAI | 32.39 | AA · 2,158 text-only questions · May 2025 revision |
| 99 | K2 Horizon 375B A23BMBZUAI Institute of Foundation Models | 31.97 | AA · 2,158 text-only questions · May 2025 revision |
| 100 | Inkling (xhigh)Thinking Machines | 31.88 | AA · 2,158 text-only questions · May 2025 revision |
| 101 | GPT-5.4 (low)OpenAI | 30.82 | AA · 2,158 text-only questions · May 2025 revision |
| 102 | Qwen3.6 Max PreviewAlibaba | 30.82 | AA · 2,158 text-only questions · May 2025 revision |
| 103 | Kimi K2.5 (Reasoning)Kimi | 30.72 | AA · 2,158 text-only questions · May 2025 revision |
| 104 | MiMo-V2-ProXiaomi | 30.44 | AA · 2,158 text-only questions · May 2025 revision |
| 105 | DeepSeek V4 Flash (Reasoning, High Effort)DeepSeek | 30.26 | AA · 2,158 text-only questions · May 2025 revision |
| 106 | Claude Opus 4.5 (Reasoning)Anthropic | 30.12 | AA · 2,158 text-only questions · May 2025 revision |
| 107 | GLM-5.1 (Reasoning)Z AI | 30.07 | AA · 2,158 text-only questions · May 2025 revision |
| 108 | Grok 4.3 (medium)SpaceXAI | 29.98 | AA · 2,158 text-only questions · May 2025 revision |
| 109 | MiniMax-M2.7MiniMax | 29.61 | AA · 2,158 text-only questions · May 2025 revision |
| 110 | A.X-K2SK Telecom | 29.56 | AA · 2,158 text-only questions · May 2025 revision |
| 111 | Gemini 3 Pro Preview (low)Google | 29.47 | AA · 2,158 text-only questions · May 2025 revision |
| 112 | GLM-5 (Reasoning)Z AI | 29.29 | AA · 2,158 text-only questions · May 2025 revision |
| 113 | Solar Pro 4Upstage | 29.24 | AA · 2,158 text-only questions · May 2025 revision |
| 114 | GPT-5.6 Terra (low)OpenAI | 29.19 | AA · 2,158 text-only questions · May 2025 revision |
| 115 | Qwen3.5 397B A17B (Reasoning)Alibaba | 28.96 | AA · 2,158 text-only questions · May 2025 revision |
| 116 | DeepSeek V3.2 SpecialeDeepSeek | 28.73 | AA · 2,158 text-only questions · May 2025 revision |
| 117 | Solar Open2 250BUpstage | 28.54 | AA · 2,158 text-only questions · May 2025 revision |
| 118 | GPT-5 (high)OpenAI | 28.5 | AA · 2,158 text-only questions · May 2025 revision |
| 119 | GPT-5.1 (high)OpenAI | 28.5 | AA · 2,158 text-only questions · May 2025 revision |
| 120 | Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA | 28.41 | AA · 2,158 text-only questions · May 2025 revision |
| 121 | GPT-5.4 nano (xhigh)OpenAI | 28.27 | AA · 2,158 text-only questions · May 2025 revision |
| 122 | GPT-5.4 mini (xhigh)OpenAI | 28.13 | AA · 2,158 text-only questions · May 2025 revision |
| 123 | Qwen3 Max ThinkingAlibaba | 27.99 | AA · 2,158 text-only questions · May 2025 revision |
| 124 | Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI | 27.94 | AA · 2,158 text-only questions · May 2025 revision |
| 125 | GLM-5.1 (Non-reasoning)Z AI | 27.94 | AA · 2,158 text-only questions · May 2025 revision |
| 126 | GPT-5 Codex (high)OpenAI | 27.85 | AA · 2,158 text-only questions · May 2025 revision |
| 127 | Qwen3.6 PlusAlibaba | 27.85 | AA · 2,158 text-only questions · May 2025 revision |
| 128 | GLM-5-TurboZ AI | 27.76 | AA · 2,158 text-only questions · May 2025 revision |
| 129 | Hy3-preview (Reasoning)Tencent | 27.76 | AA · 2,158 text-only questions · May 2025 revision |
| 130 | Grok 4.6 (low)SpaceXAI | 27.62 | AA · 2,158 text-only questions · May 2025 revision |
| 131 | GLM-4.7 (Reasoning)Z AI | 27.39 | AA · 2,158 text-only questions · May 2025 revision |
| 132 | MiMo-V2.5Xiaomi | 27.2 | AA · 2,158 text-only questions · May 2025 revision |
| 133 | Grok 4SpaceXAI | 26.69 | AA · 2,158 text-only questions · May 2025 revision |
| 134 | GPT-5.2 (medium)OpenAI | 26.69 | AA · 2,158 text-only questions · May 2025 revision |
| 135 | GPT-5.6 Luna (medium)OpenAI | 25.76 | AA · 2,158 text-only questions · May 2025 revision |
| 136 | GPT-5.1 Codex (high)OpenAI | 25.72 | AA · 2,158 text-only questions · May 2025 revision |
| 137 | GPT-5 (medium)OpenAI | 25.39 | AA · 2,158 text-only questions · May 2025 revision |
| 138 | Qwen3.5 122B A10B (Reasoning)Alibaba | 25.21 | AA · 2,158 text-only questions · May 2025 revision |
| 139 | Kimi K3 (low)Kimi | 24.98 | AA · 2,158 text-only questions · May 2025 revision |
| 140 | DeepSeek V3.2 (Reasoning)DeepSeek | 24.56 | AA · 2,158 text-only questions · May 2025 revision |
| 141 | Grok 4.20 0309 (Non-reasoning)SpaceXAI | 24.51 | AA · 2,158 text-only questions · May 2025 revision |
| 142 | Step 3.5 Flash 2603StepFun | 24.47 | AA · 2,158 text-only questions · May 2025 revision |
| 143 | Gemini 3.5 Flash (minimal)Google | 24.1 | AA · 2,158 text-only questions · May 2025 revision |
| 144 | Qwen3.5 27B (Reasoning)Alibaba | 23.91 | AA · 2,158 text-only questions · May 2025 revision |
| 145 | Kimi K2 ThinkingKimi | 23.82 | AA · 2,158 text-only questions · May 2025 revision |
| 146 | Ling 3.0 FlashInclusionAI | 23.68 | AA · 2,158 text-only questions · May 2025 revision |
| 147 | Gemma 4 31B (Reasoning)Google | 23.63 | AA · 2,158 text-only questions · May 2025 revision |
| 148 | MiniMax-M2.1MiniMax | 23.17 | AA · 2,158 text-only questions · May 2025 revision |
| 149 | Qwen3.6 27B (Reasoning)Alibaba | 23.08 | AA · 2,158 text-only questions · May 2025 revision |
| 150 | MiMo-V2-Flash (Reasoning)Xiaomi | 22.85 | AA · 2,158 text-only questions · May 2025 revision |
| 151 | MiMo-V2-Omni-0327Xiaomi | 22.57 | AA · 2,158 text-only questions · May 2025 revision |
| 152 | Gemini 2.5 ProGoogle | 22.52 | AA · 2,158 text-only questions · May 2025 revision |
| 153 | Qwen3.6 35B A3B (Reasoning)Alibaba | 22.24 | AA · 2,158 text-only questions · May 2025 revision |
| 154 | MiMo-V2-Flash (Feb 2026)Xiaomi | 22.1 | AA · 2,158 text-only questions · May 2025 revision |
| 155 | MiMo-V2-OmniXiaomi | 22.1 | AA · 2,158 text-only questions · May 2025 revision |
| 156 | Muse Glimmer (high)Meta | 21.96 | AA · 2,158 text-only questions · May 2025 revision |
| 157 | GPT-5.5 Instant (May 2026)OpenAI | 21.59 | AA · 2,158 text-only questions · May 2025 revision |
| 158 | Ring-2.6-1TInclusionAI | 21.59 | AA · 2,158 text-only questions · May 2025 revision |
| 159 | GPT-5 mini (high)OpenAI | 21.46 | AA · 2,158 text-only questions · May 2025 revision |
| 160 | Step 3.7 FlashStepFun | 21.41 | AA · 2,158 text-only questions · May 2025 revision |
| 161 | Step 3.5 FlashStepFun | 21.13 | AA · 2,158 text-only questions · May 2025 revision |
| 162 | Qwen3.5 35B A3B (Reasoning)Alibaba | 21.04 | AA · 2,158 text-only questions · May 2025 revision |
| 163 | Nemotron 3 Super 120B A12B (Reasoning)NVIDIA | 20.76 | AA · 2,158 text-only questions · May 2025 revision |
| 164 | MiniMax-M2.5MiniMax | 20.53 | AA · 2,158 text-only questions · May 2025 revision |
| 165 | o3OpenAI | 20.05 | AA · 2,158 text-only questions · May 2025 revision |
| 166 | Gemini 2.5 Pro Preview (May' 25)Google | 19.93 | AA · 2,158 text-only questions · May 2025 revision |
| 167 | GPT-5.5 Instant (June 2026)OpenAI | 19.93 | AA · 2,158 text-only questions · May 2025 revision |
| 168 | GPT-5.6 Luna (low)OpenAI | 19.83 | AA · 2,158 text-only questions · May 2025 revision |
| 169 | Qwen3.5 397B A17B (Non-reasoning)Alibaba | 19.83 | AA · 2,158 text-only questions · May 2025 revision |
| 170 | gpt-oss-120b (high)OpenAI | 19.6 | AA · 2,158 text-only questions · May 2025 revision |
| 171 | Kimi K2.6 (Non-reasoning)Kimi | 19.56 | AA · 2,158 text-only questions · May 2025 revision |
| 172 | GPT-5 (low)OpenAI | 19.56 | AA · 2,158 text-only questions · May 2025 revision |
| 173 | Grok 4.1 Fast (Reasoning)SpaceXAI | 19.32 | AA · 2,158 text-only questions · May 2025 revision |
| 174 | Gemma 4 26B A4B (Reasoning)Google | 19.32 | AA · 2,158 text-only questions · May 2025 revision |
| 175 | Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic | 19.09 | AA · 2,158 text-only questions · May 2025 revision |
| 176 | Grok 4 Fast (Reasoning)SpaceXAI | 19.09 | AA · 2,158 text-only questions · May 2025 revision |
| 177 | Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic | 19.05 | AA · 2,158 text-only questions · May 2025 revision |
| 178 | Gemini 3.5 Flash-LiteGoogle | 18.81 | AA · 2,158 text-only questions · May 2025 revision |
| 179 | Quasar 438B (max, based on GLM-5.2)Multiverse Computing | 18.67 | AA · 2,158 text-only questions · May 2025 revision |
| 180 | K-EXAONE 2.0LG AI Research | 18.58 | AA · 2,158 text-only questions · May 2025 revision |
| 181 | GPT-5.4 mini (medium)OpenAI | 18.58 | AA · 2,158 text-only questions · May 2025 revision |
| 182 | GPT-5.1 Codex mini (high)OpenAI | 18.49 | AA · 2,158 text-only questions · May 2025 revision |
| 183 | Grok 4.3 (low)SpaceXAI | 18.35 | AA · 2,158 text-only questions · May 2025 revision |
| 184 | Gemini 2.5 Pro Preview (Mar' 25)Google | 18.03 | AA · 2,158 text-only questions · May 2025 revision |
| 185 | JT-4.1 Flash 236B A21BChina Mobile | 17.79 | AA · 2,158 text-only questions · May 2025 revision |
| 186 | Claude 4.5 Sonnet (Reasoning)Anthropic | 17.79 | AA · 2,158 text-only questions · May 2025 revision |
| 187 | G9v3-39A5BAI9Stars | 17.47 | AA · 2,158 text-only questions · May 2025 revision |
| 188 | Gemini 3.1 Flash-LiteGoogle | 17.19 | AA · 2,158 text-only questions · May 2025 revision |
| 189 | Mercury 2Inception | 17.15 | AA · 2,158 text-only questions · May 2025 revision |
| 190 | GLM 5V Turbo (Reasoning)Z AI | 17.15 | AA · 2,158 text-only questions · May 2025 revision |
| 191 | GPT-5.6 Sol (Non-reasoning)OpenAI | 16.68 | AA · 2,158 text-only questions · May 2025 revision |
| 192 | HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing | 16.64 | AA · 2,158 text-only questions · May 2025 revision |
| 193 | o4-mini (high)OpenAI | 16.54 | AA · 2,158 text-only questions · May 2025 revision |
| 194 | DeepSeek V3.1 Terminus (Reasoning)DeepSeek | 16.36 | AA · 2,158 text-only questions · May 2025 revision |
| 195 | KAT Coder Pro V2KwaiKAT | 16.13 | AA · 2,158 text-only questions · May 2025 revision |
| 196 | Qwen3.5 122B A10B (Non-reasoning)Alibaba | 15.94 | AA · 2,158 text-only questions · May 2025 revision |
| 197 | GPT-5.4 nano (medium)OpenAI | 15.94 | AA · 2,158 text-only questions · May 2025 revision |
| 198 | GPT-5 mini (medium)OpenAI | 15.94 | AA · 2,158 text-only questions · May 2025 revision |
| 199 | Qwen3 235B A22B 2507 (Reasoning)Alibaba | 15.89 | AA · 2,158 text-only questions · May 2025 revision |
| 200 | DeepSeek R1 0528 (May '25)DeepSeek | 15.85 | AA · 2,158 text-only questions · May 2025 revision |
| 201 | Trinity Large ThinkingArcee AI | 15.85 | AA · 2,158 text-only questions · May 2025 revision |
| 202 | Gemma 4 12B (Reasoning)Google | 15.66 | AA · 2,158 text-only questions · May 2025 revision |
| 203 | Qwen3.6 27B (Non-reasoning)Alibaba | 15.06 | AA · 2,158 text-only questions · May 2025 revision |
| 204 | Gemini 3 Flash Preview (Non-reasoning)Google | 14.97 | AA · 2,158 text-only questions · May 2025 revision |
| 205 | Qwen3.5 9B (Reasoning)Alibaba | 14.92 | AA · 2,158 text-only questions · May 2025 revision |
| 206 | Qwen3.5 Omni PlusAlibaba | 14.87 | AA · 2,158 text-only questions · May 2025 revision |
| 207 | DeepSeek V3.2 Exp (Reasoning)DeepSeek | 14.87 | AA · 2,158 text-only questions · May 2025 revision |
| 208 | MiMo-V2.5-Pro (Non-reasoning)Xiaomi | 14.78 | AA · 2,158 text-only questions · May 2025 revision |
| 209 | GLM-4.6 (Reasoning)Z AI | 14.46 | AA · 2,158 text-only questions · May 2025 revision |
| 210 | DeepSeek V3.1 (Reasoning)DeepSeek | 14.27 | AA · 2,158 text-only questions · May 2025 revision |
| 211 | Doubao Seed CodeByteDance Seed | 14.09 | AA · 2,158 text-only questions · May 2025 revision |
| 212 | Qwen3.8 27B (medium)Alibaba | 14.09 | AA · 2,158 text-only questions · May 2025 revision |
| 213 | Qwen3.8 27B (low)Alibaba | 14.04 | AA · 2,158 text-only questions · May 2025 revision |
| 214 | Qwen3.5 27B (Non-reasoning)Alibaba | 13.95 | AA · 2,158 text-only questions · May 2025 revision |
| 215 | K-EXAONE (Reasoning)LG AI Research | 13.95 | AA · 2,158 text-only questions · May 2025 revision |
| 216 | Qwen3.6 35B A3B (Non-reasoning)Alibaba | 13.9 | AA · 2,158 text-only questions · May 2025 revision |
| 217 | Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google | 13.81 | AA · 2,158 text-only questions · May 2025 revision |
| 218 | Mistral Medium 3.5Mistral | 13.76 | AA · 2,158 text-only questions · May 2025 revision |
| 219 | GPT-5.5 (Non-reasoning)OpenAI | 13.72 | AA · 2,158 text-only questions · May 2025 revision |
| 220 | MiniMax-M2MiniMax | 13.67 | AA · 2,158 text-only questions · May 2025 revision |
| 221 | Qwen3.5 35B A3B (Non-reasoning)Alibaba | 13.44 | AA · 2,158 text-only questions · May 2025 revision |
| 222 | Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic | 13.35 | AA · 2,158 text-only questions · May 2025 revision |
| 223 | ERNIE 5.0 Thinking PreviewBaidu | 13.25 | AA · 2,158 text-only questions · May 2025 revision |
| 224 | Claude Opus 4.5 (Non-reasoning)Anthropic | 13.16 | AA · 2,158 text-only questions · May 2025 revision |
| 225 | Kimi K2.5 (Non-reasoning)Kimi | 13.16 | AA · 2,158 text-only questions · May 2025 revision |
| 226 | INTELLECT-3Prime Intellect | 13.07 | AA · 2,158 text-only questions · May 2025 revision |
| 227 | GLM-4.5 (Reasoning)Z AI | 12.97 | AA · 2,158 text-only questions · May 2025 revision |
| 228 | EXAONE 4.5 33BLG AI Research | 12.88 | AA · 2,158 text-only questions · May 2025 revision |
| 229 | Qwen3 Max Thinking (Preview)Alibaba | 12.65 | AA · 2,158 text-only questions · May 2025 revision |
| 230 | Qwen3 Next 80B A3B (Reasoning)Alibaba | 12.56 | AA · 2,158 text-only questions · May 2025 revision |
| 231 | Claude 4.1 Opus (Reasoning)Anthropic | 12.47 | AA · 2,158 text-only questions · May 2025 revision |
| 232 | Claude 4 Opus (Reasoning)Anthropic | 12.33 | AA · 2,158 text-only questions · May 2025 revision |
| 233 | Gemini 2.5 Flash (Reasoning)Google | 12.14 | AA · 2,158 text-only questions · May 2025 revision |
| 234 | Qwen3.8 27B (Non-reasoning)Alibaba | 12.14 | AA · 2,158 text-only questions · May 2025 revision |
| 235 | Apriel-v1.5-15B-ThinkerServiceNow | 12.14 | AA · 2,158 text-only questions · May 2025 revision |
| 236 | Gemini 2.5 Flash Preview (Reasoning)Google | 12.1 | AA · 2,158 text-only questions · May 2025 revision |
| 237 | o3-mini (high)OpenAI | 12.04 | AA · 2,158 text-only questions · May 2025 revision |
| 238 | Nemotron Cascade 2 30B A3BNVIDIA | 12 | AA · 2,158 text-only questions · May 2025 revision |
| 239 | Cogito v2.1 (Reasoning)Deep Cogito | 12 | AA · 2,158 text-only questions · May 2025 revision |
| 240 | Command A+Cohere | 11.96 | AA · 2,158 text-only questions · May 2025 revision |
| 241 | Qwen3 MaxAlibaba | 11.91 | AA · 2,158 text-only questions · May 2025 revision |
| 242 | Qwen3 VL 235B A22B (Reasoning)Alibaba | 11.91 | AA · 2,158 text-only questions · May 2025 revision |
| 243 | Gemma 4 31B (Non-reasoning)Google | 11.82 | AA · 2,158 text-only questions · May 2025 revision |
| 244 | Nova 2.0 Lite (high)Amazon | 11.63 | AA · 2,158 text-only questions · May 2025 revision |
| 245 | Gemma 4 26B A4B (Non-reasoning)Google | 11.49 | AA · 2,158 text-only questions · May 2025 revision |
| 246 | EXAONE 4.0 32B (Reasoning)LG AI Research | 11.45 | AA · 2,158 text-only questions · May 2025 revision |
| 247 | NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA | 11.4 | AA · 2,158 text-only questions · May 2025 revision |
| 248 | GPT-5.6 Terra (Non-reasoning)OpenAI | 11.4 | AA · 2,158 text-only questions · May 2025 revision |
| 249 | GPT-5.4 (Non-reasoning)OpenAI | 11.26 | AA · 2,158 text-only questions · May 2025 revision |
| 250 | DeepSeek V3.2 (Non-reasoning)DeepSeek | 11.21 | AA · 2,158 text-only questions · May 2025 revision |
| 251 | Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic | 11.21 | AA · 2,158 text-only questions · May 2025 revision |
| 252 | Granite 4.2 30BIBM | 11.17 | AA · 2,158 text-only questions · May 2025 revision |
| 253 | Qwen3 235B A22B 2507 InstructAlibaba | 11.12 | AA · 2,158 text-only questions · May 2025 revision |
| 254 | Ring-1TInclusionAI | 11.12 | AA · 2,158 text-only questions · May 2025 revision |
| 255 | North Mini CodeCohere | 11.08 | AA · 2,158 text-only questions · May 2025 revision |
| 256 | Sarvam 105B (high)Sarvam | 11.03 | AA · 2,158 text-only questions · May 2025 revision |
| 257 | Grok 3 mini Reasoning (high)SpaceXAI | 11.01 | AA · 2,158 text-only questions · May 2025 revision |
| 258 | gpt-oss-20b (high)OpenAI | 10.98 | AA · 2,158 text-only questions · May 2025 revision |
| 259 | Qwen3 235B A22B (Reasoning)Alibaba | 10.98 | AA · 2,158 text-only questions · May 2025 revision |
| 260 | Falcon-H1R-7BTII UAE | 10.98 | AA · 2,158 text-only questions · May 2025 revision |
| 261 | Nanbeige4.1-3BNanbeige | 10.89 | AA · 2,158 text-only questions · May 2025 revision |
| 262 | Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research | 10.89 | AA · 2,158 text-only questions · May 2025 revision |
| 263 | Step3 VL 10BStepFun | 10.84 | AA · 2,158 text-only questions · May 2025 revision |
| 264 | DiffusionGemma 26B A4BGoogle | 10.84 | AA · 2,158 text-only questions · May 2025 revision |
| 265 | Apriel-v1.6-15B-ThinkerServiceNow | 10.8 | AA · 2,158 text-only questions · May 2025 revision |
| 266 | Claude 4 Sonnet (Reasoning)Anthropic | 10.7 | AA · 2,158 text-only questions · May 2025 revision |
| 267 | Nemotron 3.5 LightningNVIDIA | 10.57 | AA · 2,158 text-only questions · May 2025 revision |
| 268 | K2-V2 (high)MBZUAI Institute of Foundation Models | 10.52 | AA · 2,158 text-only questions · May 2025 revision |
| 269 | Solar Open 100B (Reasoning)Upstage | 10.43 | AA · 2,158 text-only questions · May 2025 revision |
| 270 | Claude 4.5 Haiku (Reasoning)Anthropic | 10.38 | AA · 2,158 text-only questions · May 2025 revision |
| 271 | Qwen3 30B A3B 2507 (Reasoning)Alibaba | 10.29 | AA · 2,158 text-only questions · May 2025 revision |
| 272 | Magistral Medium 1.2Mistral | 10.29 | AA · 2,158 text-only questions · May 2025 revision |
| 273 | Solar Pro 3Upstage | 10.29 | AA · 2,158 text-only questions · May 2025 revision |
| 274 | Qwen3 Max (Preview)Alibaba | 10.15 | AA · 2,158 text-only questions · May 2025 revision |
| 275 | Qwen3 VL 32B (Reasoning)Alibaba | 10.15 | AA · 2,158 text-only questions · May 2025 revision |
| 276 | Qwen3 Coder NextAlibaba | 10.15 | AA · 2,158 text-only questions · May 2025 revision |
| 277 | K2 Think V2MBZUAI Institute of Foundation Models | 10.1 | AA · 2,158 text-only questions · May 2025 revision |
| 278 | Qwen3.5 4B (Reasoning)Alibaba | 9.92 | AA · 2,158 text-only questions · May 2025 revision |
| 279 | Mistral Small 4 (Reasoning)Mistral | 9.87 | AA · 2,158 text-only questions · May 2025 revision |
| 280 | Seed-OSS-36B-InstructByteDance Seed | 9.87 | AA · 2,158 text-only questions · May 2025 revision |
| 281 | Magistral Medium 1Mistral | 9.82 | AA · 2,158 text-only questions · May 2025 revision |
| 282 | GLM-5.2 (Non-reasoning)Z AI | 9.78 | AA · 2,158 text-only questions · May 2025 revision |
| 283 | Granite 4.2 8BIBM | 9.68 | AA · 2,158 text-only questions · May 2025 revision |
| 284 | Claude 3.7 Sonnet (Reasoning)Anthropic | 9.67 | AA · 2,158 text-only questions · May 2025 revision |
| 285 | Ring-flash-2.0InclusionAI | 9.64 | AA · 2,158 text-only questions · May 2025 revision |
| 286 | GLM-4.6V (Reasoning)Z AI | 9.64 | AA · 2,158 text-only questions · May 2025 revision |
| 287 | GPT-5 nano (high)OpenAI | 9.5 | AA · 2,158 text-only questions · May 2025 revision |
| 288 | Nova 2.0 Pro Preview (medium)Amazon | 9.41 | AA · 2,158 text-only questions · May 2025 revision |
| 289 | Qwen3.5 9B (Non-reasoning)Alibaba | 9.36 | AA · 2,158 text-only questions · May 2025 revision |
| 290 | Ling 3.0 TinyInclusionAI | 9.31 | AA · 2,158 text-only questions · May 2025 revision |
| 291 | Mi:dm K 2.5 Pro PreviewKorea Telecom | 9.08 | AA · 2,158 text-only questions · May 2025 revision |
| 292 | DeepSeek V3.2 Exp (Non-reasoning)DeepSeek | 9.04 | AA · 2,158 text-only questions · May 2025 revision |
| 293 | Nova 2.0 Lite (medium)Amazon | 8.99 | AA · 2,158 text-only questions · May 2025 revision |
| 294 | Qwen3 VL 30B A3B (Reasoning)Alibaba | 8.94 | AA · 2,158 text-only questions · May 2025 revision |
| 295 | MiniMax M1 80kMiniMax | 8.9 | AA · 2,158 text-only questions · May 2025 revision |
| 296 | Motif-2-12.7B-ReasoningMotif Technologies | 8.76 | AA · 2,158 text-only questions · May 2025 revision |
| 297 | Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research | 8.76 | AA · 2,158 text-only questions · May 2025 revision |
| 298 | Ling-2.6-1TInclusionAI | 8.71 | AA · 2,158 text-only questions · May 2025 revision |
| 299 | Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google | 8.71 | AA · 2,158 text-only questions · May 2025 revision |
| 300 | GPT-5 nano (medium)OpenAI | 8.71 | AA · 2,158 text-only questions · May 2025 revision |
| 301 | DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek | 8.67 | AA · 2,158 text-only questions · May 2025 revision |
| 302 | MiMo-V2-Flash (Non-reasoning)Xiaomi | 8.57 | AA · 2,158 text-only questions · May 2025 revision |
| 303 | DeepSeek R1 (Jan '25)DeepSeek | 8.5 | AA · 2,158 text-only questions · May 2025 revision |
| 304 | DeepSeek V4 Pro (Non-reasoning)DeepSeek | 8.25 | AA · 2,158 text-only questions · May 2025 revision |
| 305 | Mi:dm K 2.5 ProKorea Telecom | 8.06 | AA · 2,158 text-only questions · May 2025 revision |
| 306 | GPT-5.2 (Non-reasoning)OpenAI | 8.02 | AA · 2,158 text-only questions · May 2025 revision |
| 307 | Grok Code Fast 1SpaceXAI | 8.02 | AA · 2,158 text-only questions · May 2025 revision |
| 308 | Qwen3.5 4B (Non-reasoning)Alibaba | 7.97 | AA · 2,158 text-only questions · May 2025 revision |
| 309 | o3-miniOpenAI | 7.91 | AA · 2,158 text-only questions · May 2025 revision |
| 310 | MiniMax M1 40kMiniMax | 7.83 | AA · 2,158 text-only questions · May 2025 revision |
| 311 | DeepSeek V4 Flash (Non-reasoning)DeepSeek | 7.78 | AA · 2,158 text-only questions · May 2025 revision |
| 312 | Qwen3 Next 80B A3B InstructAlibaba | 7.65 | AA · 2,158 text-only questions · May 2025 revision |
| 313 | GLM-5 (Non-reasoning)Z AI | 7.65 | AA · 2,158 text-only questions · May 2025 revision |
| 314 | GLM-4.7-Flash (Reasoning)Z AI | 7.6 | AA · 2,158 text-only questions · May 2025 revision |
| 315 | Qwen3.5 Omni FlashAlibaba | 7.6 | AA · 2,158 text-only questions · May 2025 revision |
| 316 | Magistral Small 1Mistral | 7.55 | AA · 2,158 text-only questions · May 2025 revision |
| 317 | Sarvam 30B (high)Sarvam | 7.51 | AA · 2,158 text-only questions · May 2025 revision |
| 318 | Qwen3 Omni 30B A3B (Reasoning)Alibaba | 7.46 | AA · 2,158 text-only questions · May 2025 revision |
| 319 | Kimi K2Kimi | 7.41 | AA · 2,158 text-only questions · May 2025 revision |
| 320 | Qwen3 32B (Reasoning)Alibaba | 7.41 | AA · 2,158 text-only questions · May 2025 revision |
| 321 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA | 7.38 | AA · 2,158 text-only questions · May 2025 revision |
| 322 | Solar Pro 2 (Reasoning)Upstage | 7.37 | AA · 2,158 text-only questions · May 2025 revision |
| 323 | QwQ 32BAlibaba | 7.35 | AA · 2,158 text-only questions · May 2025 revision |
| 324 | Ling-1TInclusionAI | 7.32 | AA · 2,158 text-only questions · May 2025 revision |
| 325 | Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA | 7.32 | AA · 2,158 text-only questions · May 2025 revision |
| 326 | Claude 4.5 Sonnet (Non-reasoning)Anthropic | 7.23 | AA · 2,158 text-only questions · May 2025 revision |
| 327 | GPT-5.6 Luna (Non-reasoning)OpenAI | 7.23 | AA · 2,158 text-only questions · May 2025 revision |
| 328 | GLM-4.5-AirZ AI | 7.04 | AA · 2,158 text-only questions · May 2025 revision |
| 329 | o1OpenAI | 7.01 | AA · 2,158 text-only questions · May 2025 revision |
| 330 | Nova 2.0 Omni (medium)Amazon | 6.95 | AA · 2,158 text-only questions · May 2025 revision |
| 331 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google | 6.95 | AA · 2,158 text-only questions · May 2025 revision |
| 332 | Hy3-preview (Non-reasoning)Tencent | 6.95 | AA · 2,158 text-only questions · May 2025 revision |
| 333 | Qwen3 30B A3B 2507 InstructAlibaba | 6.9 | AA · 2,158 text-only questions · May 2025 revision |
| 334 | LFM2.5-8B-A1BLiquid AI | 6.86 | AA · 2,158 text-only questions · May 2025 revision |
| 335 | Qwen3 VL 32B InstructAlibaba | 6.81 | AA · 2,158 text-only questions · May 2025 revision |
| 336 | Celeris-1Celeris | 6.81 | AA · 2,158 text-only questions · May 2025 revision |
| 337 | Gemini 2.5 Flash-Lite (Reasoning)Google | 6.81 | AA · 2,158 text-only questions · May 2025 revision |
| 338 | Grok 4.3 (Non-reasoning)SpaceXAI | 6.77 | AA · 2,158 text-only questions · May 2025 revision |
| 339 | LFM2.5-1.2B-InstructLiquid AI | 6.72 | AA · 2,158 text-only questions · May 2025 revision |
| 340 | DeepSeek V3.1 (Non-reasoning)DeepSeek | 6.67 | AA · 2,158 text-only questions · May 2025 revision |
| 341 | Granite 4.2 3BIBM | 6.63 | AA · 2,158 text-only questions · May 2025 revision |
| 342 | Qwen3 VL 235B A22B InstructAlibaba | 6.63 | AA · 2,158 text-only questions · May 2025 revision |
| 343 | GPT-5 (ChatGPT)OpenAI | 6.63 | AA · 2,158 text-only questions · May 2025 revision |
| 344 | Sonar ProPerplexity | 6.55 | AA · 2,158 text-only questions · May 2025 revision |
| 345 | MiniCPM5-1B (Reasoning)OpenBMB | 6.49 | AA · 2,158 text-only questions · May 2025 revision |
| 346 | JT-35B-FlashChina Mobile | 6.44 | AA · 2,158 text-only questions · May 2025 revision |
| 347 | JT-MINIChina Mobile | 6.44 | AA · 2,158 text-only questions · May 2025 revision |
| 348 | Magistral Small 1.2Mistral | 6.44 | AA · 2,158 text-only questions · May 2025 revision |
| 349 | Granite 4.0 H 350MIBM | 6.44 | AA · 2,158 text-only questions · May 2025 revision |
| 350 | Olmo 3 32B ThinkAllen Institute for AI | 6.44 | AA · 2,158 text-only questions · May 2025 revision |
| 351 | Kimi K2 0905Kimi | 6.39 | AA · 2,158 text-only questions · May 2025 revision |
| 352 | GLM-4.7 (Non-reasoning)Z AI | 6.39 | AA · 2,158 text-only questions · May 2025 revision |
| 353 | Gemma 4 12B (Non-reasoning)Google | 6.35 | AA · 2,158 text-only questions · May 2025 revision |
| 354 | Gemini 2.0 Flash Thinking Experimental (Jan '25)Google | 6.35 | AA · 2,158 text-only questions · May 2025 revision |
| 355 | GLM-4.5V (Reasoning)Z AI | 6.3 | AA · 2,158 text-only questions · May 2025 revision |
| 356 | Qwen3 VL 30B A3B InstructAlibaba | 6.3 | AA · 2,158 text-only questions · May 2025 revision |
| 357 | Olmo 3.1 32B ThinkAllen Institute for AI | 6.26 | AA · 2,158 text-only questions · May 2025 revision |
| 358 | Ling 2.6 FlashInclusionAI | 6.26 | AA · 2,158 text-only questions · May 2025 revision |
| 359 | LFM2.5-1.2B-ThinkingLiquid AI | 6.21 | AA · 2,158 text-only questions · May 2025 revision |
| 360 | Ling-flash-2.0InclusionAI | 6.21 | AA · 2,158 text-only questions · May 2025 revision |
| 361 | LFM2.5-2.6BLiquid AI | 6.21 | AA · 2,158 text-only questions · May 2025 revision |
| 362 | Gemini 2.0 Pro Experimental (Feb '25)Google | 6.19 | AA · 2,158 text-only questions · May 2025 revision |
| 363 | Qwen3 4B 2507 (Reasoning)Alibaba | 6.16 | AA · 2,158 text-only questions · May 2025 revision |
| 364 | Qwen3 30B A3B (Reasoning)Alibaba | 6.16 | AA · 2,158 text-only questions · May 2025 revision |
| 365 | Claude 4 Opus (Non-reasoning)Anthropic | 6.16 | AA · 2,158 text-only questions · May 2025 revision |
| 366 | Solar Pro 2 (Preview) (Reasoning)Upstage | 6.07 | AA · 2,158 text-only questions · May 2025 revision |
| 367 | GPT-5 (minimal)OpenAI | 6.02 | AA · 2,158 text-only questions · May 2025 revision |
| 368 | Olmo 3 7B ThinkAllen Institute for AI | 5.98 | AA · 2,158 text-only questions · May 2025 revision |
| 369 | Exaone 4.0 1.2B (Reasoning)LG AI Research | 5.98 | AA · 2,158 text-only questions · May 2025 revision |
| 370 | Tri-21B-ThinkTrillion Labs | 5.93 | AA · 2,158 text-only questions · May 2025 revision |
| 371 | DeepSeek R1 0528 Qwen3 8BDeepSeek | 5.93 | AA · 2,158 text-only questions · May 2025 revision |
| 372 | GPT-5.4 mini (Non-Reasoning)OpenAI | 5.89 | AA · 2,158 text-only questions · May 2025 revision |
| 373 | gpt-oss-120b (low)OpenAI | 5.89 | AA · 2,158 text-only questions · May 2025 revision |
| 374 | LongCat Flash LiteLongCat | 5.84 | AA · 2,158 text-only questions · May 2025 revision |
| 375 | Tri-21B-think PreviewTrillion Labs | 5.79 | AA · 2,158 text-only questions · May 2025 revision |
| 376 | Olmo 3 7B InstructAllen Institute for AI | 5.79 | AA · 2,158 text-only questions · May 2025 revision |
| 377 | Exaone 4.0 1.2B (Non-reasoning)LG AI Research | 5.7 | AA · 2,158 text-only questions · May 2025 revision |
| 378 | K-EXAONE (Non-reasoning)LG AI Research | 5.7 | AA · 2,158 text-only questions · May 2025 revision |
| 379 | Qwen3 0.6B (Reasoning)Alibaba | 5.64 | AA · 2,158 text-only questions · May 2025 revision |
| 380 | Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIA | 5.62 | AA · 2,158 text-only questions · May 2025 revision |
| 381 | LFM2 1.2BLiquid AI | 5.56 | AA · 2,158 text-only questions · May 2025 revision |
| 382 | Llama 3.2 Instruct 11B (Vision)Meta | 5.53 | AA · 2,158 text-only questions · May 2025 revision |
| 383 | Granite 4.0 350MIBM | 5.51 | AA · 2,158 text-only questions · May 2025 revision |
| 384 | NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA | 5.51 | AA · 2,158 text-only questions · May 2025 revision |
| 385 | HyperCLOVA X SEED Think (32B)Naver | 5.47 | AA · 2,158 text-only questions · May 2025 revision |
| 386 | Apertus 70B InstructSwiss AI Initiative | 5.47 | AA · 2,158 text-only questions · May 2025 revision |
| 387 | LFM2 2.6BLiquid AI | 5.47 | AA · 2,158 text-only questions · May 2025 revision |
| 388 | GLM-4.6 (Non-reasoning)Z AI | 5.47 | AA · 2,158 text-only questions · May 2025 revision |
| 389 | Llama 3.2 Instruct 1BMeta | 5.46 | AA · 2,158 text-only questions · May 2025 revision |
| 390 | Ministral 3 3BMistral | 5.38 | AA · 2,158 text-only questions · May 2025 revision |
| 391 | OLMo 2 7BAllen Institute for AI | 5.38 | AA · 2,158 text-only questions · May 2025 revision |
| 392 | DeepSeek Coder V2 Lite InstructDeepSeek | 5.36 | AA · 2,158 text-only questions · May 2025 revision |
| 393 | Llama 3.2 Instruct 3BMeta | 5.35 | AA · 2,158 text-only questions · May 2025 revision |
| 394 | Gemma 3 1B InstructGoogle | 5.33 | AA · 2,158 text-only questions · May 2025 revision |
| 395 | Llama 2 Chat 7BMeta | 5.33 | AA · 2,158 text-only questions · May 2025 revision |
| 396 | Qwen3 1.7B (Non-reasoning)Alibaba | 5.32 | AA · 2,158 text-only questions · May 2025 revision |
| 397 | Gemma 3 4B InstructGoogle | 5.29 | AA · 2,158 text-only questions · May 2025 revision |
| 398 | Llama 3.1 Instruct 8BMeta | 5.29 | AA · 2,158 text-only questions · May 2025 revision |
| 399 | gpt-oss-20b (low)OpenAI | 5.28 | AA · 2,158 text-only questions · May 2025 revision |
| 400 | GPT-5.1 (Non-reasoning)OpenAI | 5.28 | AA · 2,158 text-only questions · May 2025 revision |
| 401 | Tiny Aya GlobalCohere | 5.24 | AA · 2,158 text-only questions · May 2025 revision |
| 402 | Llama 2 Chat 70BMeta | 5.21 | AA · 2,158 text-only questions · May 2025 revision |
| 403 | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIA | 5.19 | AA · 2,158 text-only questions · May 2025 revision |
| 404 | Nova 2.0 Pro Preview (low)Amazon | 5.19 | AA · 2,158 text-only questions · May 2025 revision |
| 405 | Jamba 1.5 MiniAI21 Labs | 5.14 | AA · 2,158 text-only questions · May 2025 revision |
| 406 | Molmo 7B-DAllen Institute for AI | 5.14 | AA · 2,158 text-only questions · May 2025 revision |
| 407 | DeepSeek R1 Distill Llama 70BDeepSeek | 5.13 | AA · 2,158 text-only questions · May 2025 revision |
| 408 | LFM2.5-VL-1.6BLiquid AI | 5.1 | AA · 2,158 text-only questions · May 2025 revision |
| 409 | Qwen3.5 0.8B (Non-reasoning)Alibaba | 5.1 | AA · 2,158 text-only questions · May 2025 revision |
| 410 | Grok 4.1 Fast (Non-reasoning)SpaceXAI | 5.1 | AA · 2,158 text-only questions · May 2025 revision |
| 411 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google | 5.1 | AA · 2,158 text-only questions · May 2025 revision |
| 412 | Llama 3 Instruct 8BMeta | 5.06 | AA · 2,158 text-only questions · May 2025 revision |
| 413 | Ling-mini-2.0InclusionAI | 5.05 | AA · 2,158 text-only questions · May 2025 revision |
| 414 | MiniCPM-V 4.6 1.3BOpenBMB | 5.05 | AA · 2,158 text-only questions · May 2025 revision |
| 415 | GPT-5 mini (minimal)OpenAI | 5.05 | AA · 2,158 text-only questions · May 2025 revision |
| 416 | GPT-4.1 miniOpenAI | 5.02 | AA · 2,158 text-only questions · May 2025 revision |
| 417 | Granite 4.0 MicroIBM | 5 | AA · 2,158 text-only questions · May 2025 revision |
| 418 | Phi-4 Multimodal InstructMicrosoft | 5 | AA · 2,158 text-only questions · May 2025 revision |
| 419 | EXAONE 4.0 32B (Non-reasoning)LG AI Research | 5 | AA · 2,158 text-only questions · May 2025 revision |
| 420 | Olmo 3.1 32B InstructAllen Institute for AI | 5 | AA · 2,158 text-only questions · May 2025 revision |
| 421 | Granite 4.0 H 1BIBM | 5 | AA · 2,158 text-only questions · May 2025 revision |
| 422 | Qwen3.5 2B (Non-reasoning)Alibaba | 4.96 | AA · 2,158 text-only questions · May 2025 revision |
| 423 | Apertus 8B InstructSwiss AI Initiative | 4.96 | AA · 2,158 text-only questions · May 2025 revision |
| 424 | GLM-4.7-Flash (Non-reasoning)Z AI | 4.96 | AA · 2,158 text-only questions · May 2025 revision |
| 425 | Phi-3 Mini Instruct 3.8BMicrosoft | 4.96 | AA · 2,158 text-only questions · May 2025 revision |
| 426 | Qwen3 0.6B (Non-reasoning)Alibaba | 4.93 | AA · 2,158 text-only questions · May 2025 revision |
| 427 | LFM 40BLiquid AI | 4.93 | AA · 2,158 text-only questions · May 2025 revision |
| 428 | LFM2 8B A1BLiquid AI | 4.91 | AA · 2,158 text-only questions · May 2025 revision |
| 429 | Llama 4 MaverickMeta | 4.91 | AA · 2,158 text-only questions · May 2025 revision |
| 430 | Qwen2.5 Coder Instruct 7BAlibaba | 4.89 | AA · 2,158 text-only questions · May 2025 revision |
| 431 | SonarPerplexity | 4.88 | AA · 2,158 text-only questions · May 2025 revision |
| 432 | NVIDIA Nemotron 3 Nano 4BNVIDIA | 4.87 | AA · 2,158 text-only questions · May 2025 revision |
| 433 | NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA | 4.87 | AA · 2,158 text-only questions · May 2025 revision |
| 434 | Gemma 4 E4B (Non-reasoning)Google | 4.82 | AA · 2,158 text-only questions · May 2025 revision |
| 435 | Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA | 4.82 | AA · 2,158 text-only questions · May 2025 revision |
| 436 | Granite 4.0 1BIBM | 4.82 | AA · 2,158 text-only questions · May 2025 revision |
| 437 | Gemma 3n E4B Instruct Preview (May '25)Google | 4.82 | AA · 2,158 text-only questions · May 2025 revision |
| 438 | Command-R (Mar '24)Cohere | 4.79 | AA · 2,158 text-only questions · May 2025 revision |
| 439 | OpenChat 3.5 (1210)OpenChat | 4.79 | AA · 2,158 text-only questions · May 2025 revision |
| 440 | Llama 2 Chat 13BMeta | 4.77 | AA · 2,158 text-only questions · May 2025 revision |
| 441 | Gemma 4 E2B (Reasoning)Google | 4.77 | AA · 2,158 text-only questions · May 2025 revision |
| 442 | DeepSeek V3 0324DeepSeek | 4.74 | AA · 2,158 text-only questions · May 2025 revision |
| 443 | Gemini 2.5 Flash (Non-reasoning)Google | 4.73 | AA · 2,158 text-only questions · May 2025 revision |
| 444 | Gemma 4 E2B (Non-reasoning)Google | 4.73 | AA · 2,158 text-only questions · May 2025 revision |
| 445 | Gemini 1.5 Flash-8BGoogle | 4.69 | AA · 2,158 text-only questions · May 2025 revision |
| 446 | Mistral Medium 3.1Mistral | 4.68 | AA · 2,158 text-only questions · May 2025 revision |
| 447 | Mixtral 8x7B InstructMistral | 4.68 | AA · 2,158 text-only questions · May 2025 revision |
| 448 | Gemini 1.5 Pro (Sep '24)Google | 4.64 | AA · 2,158 text-only questions · May 2025 revision |
| 449 | Qwen3 1.7B (Reasoning)Alibaba | 4.64 | AA · 2,158 text-only questions · May 2025 revision |
| 450 | Qwen3 Omni 30B A3B InstructAlibaba | 4.63 | AA · 2,158 text-only questions · May 2025 revision |
| 451 | Nova MicroAmazon | 4.63 | AA · 2,158 text-only questions · May 2025 revision |
| 452 | Ministral 3 14BMistral | 4.63 | AA · 2,158 text-only questions · May 2025 revision |
| 453 | Command-R+ (Apr '24)Cohere | 4.63 | AA · 2,158 text-only questions · May 2025 revision |
| 454 | Qwen3 30B A3B (Non-reasoning)Alibaba | 4.6 | AA · 2,158 text-only questions · May 2025 revision |
| 455 | NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA | 4.59 | AA · 2,158 text-only questions · May 2025 revision |
| 456 | NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA | 4.59 | AA · 2,158 text-only questions · May 2025 revision |
| 457 | Qwen3 VL 4B (Reasoning)Alibaba | 4.59 | AA · 2,158 text-only questions · May 2025 revision |
| 458 | DeepSeek R1 Distill Qwen 32BDeepSeek | 4.58 | AA · 2,158 text-only questions · May 2025 revision |
| 459 | Mistral 7B InstructMistral | 4.55 | AA · 2,158 text-only questions · May 2025 revision |
| 460 | Qwen3 Coder 480B A35B InstructAlibaba | 4.54 | AA · 2,158 text-only questions · May 2025 revision |
| 461 | Grok 4 Fast (Non-reasoning)SpaceXAI | 4.54 | AA · 2,158 text-only questions · May 2025 revision |
| 462 | MiniCPM5-1B (Non-reasoning)OpenBMB | 4.54 | AA · 2,158 text-only questions · May 2025 revision |
| 463 | Qwen3 14B (Reasoning)Alibaba | 4.53 | AA · 2,158 text-only questions · May 2025 revision |
| 464 | Llama 3 Instruct 70BMeta | 4.52 | AA · 2,158 text-only questions · May 2025 revision |
| 465 | G9v3-3BAI9Stars | 4.49 | AA · 2,158 text-only questions · May 2025 revision |
| 466 | Llama 3.2 Instruct 90B (Vision)Meta | 4.49 | AA · 2,158 text-only questions · May 2025 revision |
| 467 | Gemma 3n E4B InstructGoogle | 4.49 | AA · 2,158 text-only questions · May 2025 revision |
| 468 | Qwen3 4B 2507 InstructAlibaba | 4.49 | AA · 2,158 text-only questions · May 2025 revision |
| 469 | Llama 3.1 Instruct 70BMeta | 4.47 | AA · 2,158 text-only questions · May 2025 revision |
| 470 | Phi-4 Mini InstructMicrosoft | 4.45 | AA · 2,158 text-only questions · May 2025 revision |
| 471 | Grok BetaSpaceXAI | 4.45 | AA · 2,158 text-only questions · May 2025 revision |
| 472 | Jamba 1.7 MiniAI21 Labs | 4.45 | AA · 2,158 text-only questions · May 2025 revision |
| 473 | Reka Flash 3Reka AI | 4.44 | AA · 2,158 text-only questions · May 2025 revision |
| 474 | Qwen3 4B (Reasoning)Alibaba | 4.42 | AA · 2,158 text-only questions · May 2025 revision |
| 475 | Gemma 3 27B InstructGoogle | 4.4 | AA · 2,158 text-only questions · May 2025 revision |
| 476 | Gemini 1.5 Flash (May '24)Google | 4.36 | AA · 2,158 text-only questions · May 2025 revision |
| 477 | K2-V2 (medium)MBZUAI Institute of Foundation Models | 4.36 | AA · 2,158 text-only questions · May 2025 revision |
| 478 | Mistral Small 3.1Mistral | 4.31 | AA · 2,158 text-only questions · May 2025 revision |
| 479 | Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA | 4.31 | AA · 2,158 text-only questions · May 2025 revision |
| 480 | Mistral Small 3.2Mistral | 4.31 | AA · 2,158 text-only questions · May 2025 revision |
| 481 | Jamba 1.6 MiniAI21 Labs | 4.31 | AA · 2,158 text-only questions · May 2025 revision |
| 482 | Mistral SabaMistral | 4.3 | AA · 2,158 text-only questions · May 2025 revision |
| 483 | Nova LiteAmazon | 4.29 | AA · 2,158 text-only questions · May 2025 revision |
| 484 | Molmo2-8BAllen Institute for AI | 4.26 | AA · 2,158 text-only questions · May 2025 revision |
| 485 | Ministral 3 8BMistral | 4.26 | AA · 2,158 text-only questions · May 2025 revision |
| 486 | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA | 4.26 | AA · 2,158 text-only questions · May 2025 revision |
| 487 | DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research | 4.26 | AA · 2,158 text-only questions · May 2025 revision |
| 488 | Gemini 2.0 Flash (Feb '25)Google | 4.26 | AA · 2,158 text-only questions · May 2025 revision |
| 489 | Claude 4 Sonnet (Non-reasoning)Anthropic | 4.26 | AA · 2,158 text-only questions · May 2025 revision |
| 490 | Mistral Small (Sep '24)Mistral | 4.25 | AA · 2,158 text-only questions · May 2025 revision |
| 491 | Claude 4.5 Haiku (Non-reasoning)Anthropic | 4.22 | AA · 2,158 text-only questions · May 2025 revision |
| 492 | Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research | 4.22 | AA · 2,158 text-only questions · May 2025 revision |
| 493 | LFM2 24B A2BLiquid AI | 4.22 | AA · 2,158 text-only questions · May 2025 revision |
| 494 | Gemma 3 12B InstructGoogle | 4.21 | AA · 2,158 text-only questions · May 2025 revision |
| 495 | GPT-4o miniOpenAI | 4.2 | AA · 2,158 text-only questions · May 2025 revision |
| 496 | Qwen3 235B A22B (Non-reasoning)Alibaba | 4.19 | AA · 2,158 text-only questions · May 2025 revision |
| 497 | Llama 3.1 Nemotron Instruct 70BNVIDIA | 4.19 | AA · 2,158 text-only questions · May 2025 revision |
| 498 | Gemini 1.0 ProGoogle | 4.19 | AA · 2,158 text-only questions · May 2025 revision |
| 499 | GPT-4.1OpenAI | 4.18 | AA · 2,158 text-only questions · May 2025 revision |
| 500 | Granite 3.3 8B (Non-reasoning)IBM | 4.17 | AA · 2,158 text-only questions · May 2025 revision |
| 501 | Mistral Large 3Mistral | 4.17 | AA · 2,158 text-only questions · May 2025 revision |
| 502 | Gemma 3n E2B InstructGoogle | 4.17 | AA · 2,158 text-only questions · May 2025 revision |
| 503 | Nova PremierAmazon | 4.17 | AA · 2,158 text-only questions · May 2025 revision |
| 504 | Claude 3.7 Sonnet (Non-reasoning)Anthropic | 4.16 | AA · 2,158 text-only questions · May 2025 revision |
| 505 | DeepSeek R1 Distill Qwen 14BDeepSeek | 4.13 | AA · 2,158 text-only questions · May 2025 revision |
| 506 | Qwen3 32B (Non-reasoning)Alibaba | 4.13 | AA · 2,158 text-only questions · May 2025 revision |
| 507 | Grok 3SpaceXAI | 4.12 | AA · 2,158 text-only questions · May 2025 revision |
| 508 | Granite 4.1 30BIBM | 4.12 | AA · 2,158 text-only questions · May 2025 revision |
| 509 | Qwen2.5 TurboAlibaba | 4.09 | AA · 2,158 text-only questions · May 2025 revision |
| 510 | Mistral Small (Feb '24)Mistral | 4.09 | AA · 2,158 text-only questions · May 2025 revision |
| 511 | Gemini 2.0 Flash-Lite (Preview)Google | 4.09 | AA · 2,158 text-only questions · May 2025 revision |
| 512 | Qwen3 14B (Non-reasoning)Alibaba | 4.09 | AA · 2,158 text-only questions · May 2025 revision |
| 513 | Claude 3 HaikuAnthropic | 4.08 | AA · 2,158 text-only questions · May 2025 revision |
| 514 | GPT-5.4 nano (Non-Reasoning)OpenAI | 4.08 | AA · 2,158 text-only questions · May 2025 revision |
| 515 | Jamba 1.5 LargeAI21 Labs | 4.07 | AA · 2,158 text-only questions · May 2025 revision |
| 516 | Gemini 2.0 Flash (experimental)Google | 4.07 | AA · 2,158 text-only questions · May 2025 revision |
| 517 | Mistral Medium 3Mistral | 4.05 | AA · 2,158 text-only questions · May 2025 revision |
| 518 | Command ACohere | 4.03 | AA · 2,158 text-only questions · May 2025 revision |
| 519 | Nova 2.0 Lite (low)Amazon | 4.03 | AA · 2,158 text-only questions · May 2025 revision |
| 520 | Devstral Small (May '25)Mistral | 4.03 | AA · 2,158 text-only questions · May 2025 revision |
| 521 | Mixtral 8x22B InstructMistral | 4.01 | AA · 2,158 text-only questions · May 2025 revision |
| 522 | GPT-5 nano (minimal)OpenAI | 3.99 | AA · 2,158 text-only questions · May 2025 revision |
| 523 | Qwen2.5 Instruct 32BAlibaba | 3.99 | AA · 2,158 text-only questions · May 2025 revision |
| 524 | Hermes 3 - Llama-3.1 70BNous Research | 3.99 | AA · 2,158 text-only questions · May 2025 revision |
| 525 | Llama 3.1 Instruct 405BMeta | 3.98 | AA · 2,158 text-only questions · May 2025 revision |
| 526 | GPT-4o (March 2025, chatgpt-4o-latest)OpenAI | 3.97 | AA · 2,158 text-only questions · May 2025 revision |
| 527 | Nova 2.0 Pro Preview (Non-reasoning)Amazon | 3.94 | AA · 2,158 text-only questions · May 2025 revision |
| 528 | QwQ 32B-PreviewAlibaba | 3.91 | AA · 2,158 text-only questions · May 2025 revision |
| 529 | Qwen3 8B (Reasoning)Alibaba | 3.89 | AA · 2,158 text-only questions · May 2025 revision |
| 530 | Claude 2.1Anthropic | 3.89 | AA · 2,158 text-only questions · May 2025 revision |
| 531 | Nova 2.0 Omni (Non-reasoning)Amazon | 3.85 | AA · 2,158 text-only questions · May 2025 revision |
| 532 | Qwen3 Coder 30B A3B InstructAlibaba | 3.85 | AA · 2,158 text-only questions · May 2025 revision |
| 533 | Qwen3 VL 8B (Reasoning)Alibaba | 3.85 | AA · 2,158 text-only questions · May 2025 revision |
| 534 | Qwen2.5 MaxAlibaba | 3.82 | AA · 2,158 text-only questions · May 2025 revision |
| 535 | Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA | 3.81 | AA · 2,158 text-only questions · May 2025 revision |
| 536 | Mistral Small 4 (Non-reasoning)Mistral | 3.8 | AA · 2,158 text-only questions · May 2025 revision |
| 537 | Gemini 2.5 Flash Preview (Non-reasoning)Google | 3.8 | AA · 2,158 text-only questions · May 2025 revision |
| 538 | Granite 4.0 H SmallIBM | 3.8 | AA · 2,158 text-only questions · May 2025 revision |
| 539 | Granite 4.1 8BIBM | 3.8 | AA · 2,158 text-only questions · May 2025 revision |
| 540 | Llama 4 ScoutMeta | 3.78 | AA · 2,158 text-only questions · May 2025 revision |
| 541 | DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research | 3.76 | AA · 2,158 text-only questions · May 2025 revision |
| 542 | Mistral Small 3Mistral | 3.76 | AA · 2,158 text-only questions · May 2025 revision |
| 543 | Phi-4Microsoft | 3.76 | AA · 2,158 text-only questions · May 2025 revision |
| 544 | GPT-4.1 nanoOpenAI | 3.75 | AA · 2,158 text-only questions · May 2025 revision |
| 545 | Devstral Small (Jul '25)Mistral | 3.75 | AA · 2,158 text-only questions · May 2025 revision |
| 546 | Devstral MediumMistral | 3.75 | AA · 2,158 text-only questions · May 2025 revision |
| 547 | Gemma 4 E4B (Reasoning)Google | 3.75 | AA · 2,158 text-only questions · May 2025 revision |
| 548 | Jamba Reasoning 3BAI21 Labs | 3.75 | AA · 2,158 text-only questions · May 2025 revision |
| 549 | Solar Pro 2 (Non-reasoning)Upstage | 3.71 | AA · 2,158 text-only questions · May 2025 revision |
| 550 | Claude 3.5 Sonnet (Oct '24)Anthropic | 3.69 | AA · 2,158 text-only questions · May 2025 revision |
| 551 | Gemini 2.5 Flash-Lite (Non-reasoning)Google | 3.66 | AA · 2,158 text-only questions · May 2025 revision |
| 552 | GLM-4.6V (Non-reasoning)Z AI | 3.66 | AA · 2,158 text-only questions · May 2025 revision |
| 553 | Solar Pro 2 (Preview) (Non-reasoning)Upstage | 3.66 | AA · 2,158 text-only questions · May 2025 revision |
| 554 | Qwen2 Instruct 72BAlibaba | 3.66 | AA · 2,158 text-only questions · May 2025 revision |
| 555 | Jamba 1.7 LargeAI21 Labs | 3.66 | AA · 2,158 text-only questions · May 2025 revision |
| 556 | Claude 3.5 HaikuAnthropic | 3.63 | AA · 2,158 text-only questions · May 2025 revision |
| 557 | Claude 3 SonnetAnthropic | 3.62 | AA · 2,158 text-only questions · May 2025 revision |
| 558 | Gemma 3 270MGoogle | 3.61 | AA · 2,158 text-only questions · May 2025 revision |
| 559 | OLMo 2 32BAllen Institute for AI | 3.61 | AA · 2,158 text-only questions · May 2025 revision |
| 560 | Jamba 1.6 LargeAI21 Labs | 3.61 | AA · 2,158 text-only questions · May 2025 revision |
| 561 | K2-V2 (low)MBZUAI Institute of Foundation Models | 3.61 | AA · 2,158 text-only questions · May 2025 revision |
| 562 | Qwen3 VL 4B InstructAlibaba | 3.61 | AA · 2,158 text-only questions · May 2025 revision |
| 563 | Devstral 2Mistral | 3.57 | AA · 2,158 text-only questions · May 2025 revision |
| 564 | Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research | 3.57 | AA · 2,158 text-only questions · May 2025 revision |
| 565 | o1-miniOpenAI | 3.57 | AA · 2,158 text-only questions · May 2025 revision |
| 566 | Llama 3.3 Instruct 70BMeta | 3.56 | AA · 2,158 text-only questions · May 2025 revision |
| 567 | Qwen2.5 Instruct 72BAlibaba | 3.56 | AA · 2,158 text-only questions · May 2025 revision |
| 568 | Claude InstantAnthropic | 3.53 | AA · 2,158 text-only questions · May 2025 revision |
| 569 | Qwen2.5 Coder Instruct 32BAlibaba | 3.51 | AA · 2,158 text-only questions · May 2025 revision |
| 570 | Mistral MediumMistral | 3.51 | AA · 2,158 text-only questions · May 2025 revision |
| 571 | Mistral Large (Feb '24)Mistral | 3.5 | AA · 2,158 text-only questions · May 2025 revision |
| 572 | GLM-4.5V (Non-reasoning)Z AI | 3.48 | AA · 2,158 text-only questions · May 2025 revision |
| 573 | Gemini 1.5 Pro (May '24)Google | 3.48 | AA · 2,158 text-only questions · May 2025 revision |
| 574 | Devstral Small 2Mistral | 3.48 | AA · 2,158 text-only questions · May 2025 revision |
| 575 | Granite 4.1 3BIBM | 3.43 | AA · 2,158 text-only questions · May 2025 revision |
| 576 | Gemini 2.0 Flash-Lite (Feb '25)Google | 3.35 | AA · 2,158 text-only questions · May 2025 revision |
| 577 | Mistral Large 2 (Nov '24)Mistral | 3.34 | AA · 2,158 text-only questions · May 2025 revision |
| 578 | ERNIE 4.5 300B A47BBaidu | 3.34 | AA · 2,158 text-only questions · May 2025 revision |
| 579 | Qwen3 4B (Non-reasoning)Alibaba | 3.32 | AA · 2,158 text-only questions · May 2025 revision |
| 580 | Llama 3.1 Tulu3 405BAllen Institute for AI | 3.28 | AA · 2,158 text-only questions · May 2025 revision |
| 581 | Claude 3.5 Sonnet (June '24)Anthropic | 3.23 | AA · 2,158 text-only questions · May 2025 revision |
| 582 | Nova ProAmazon | 3.19 | AA · 2,158 text-only questions · May 2025 revision |
| 583 | Gemini 1.5 Flash (Sep '24)Google | 3.17 | AA · 2,158 text-only questions · May 2025 revision |
| 584 | GPT-4 TurboOpenAI | 3.13 | AA · 2,158 text-only questions · May 2025 revision |
| 585 | Sarvam M (Reasoning)Sarvam | 3.1 | AA · 2,158 text-only questions · May 2025 revision |
| 586 | DeepSeek R1 Distill Qwen 1.5BDeepSeek | 3.08 | AA · 2,158 text-only questions · May 2025 revision |
| 587 | Grok 2 (Dec '24)SpaceXAI | 3.06 | AA · 2,158 text-only questions · May 2025 revision |
| 588 | Mistral Large 2 (Jul '24)Mistral | 2.95 | AA · 2,158 text-only questions · May 2025 revision |
| 589 | DBRX InstructDatabricks | 2.92 | AA · 2,158 text-only questions · May 2025 revision |
| 590 | DeepSeek V3 (Dec '24)DeepSeek | 2.9 | AA · 2,158 text-only questions · May 2025 revision |
| 591 | Nova 2.0 Lite (Non-reasoning)Amazon | 2.87 | AA · 2,158 text-only questions · May 2025 revision |
| 592 | Pixtral LargeMistral | 2.8 | AA · 2,158 text-only questions · May 2025 revision |
| 593 | Claude 3 OpusAnthropic | 2.75 | AA · 2,158 text-only questions · May 2025 revision |
| 594 | GPT-4o (ChatGPT)OpenAI | 2.73 | AA · 2,158 text-only questions · May 2025 revision |
| 595 | Qwen3 VL 8B InstructAlibaba | 2.69 | AA · 2,158 text-only questions · May 2025 revision |
| 596 | Qwen3.5 2B (Reasoning)Alibaba | 2.59 | AA · 2,158 text-only questions · May 2025 revision |
| 597 | Kimi Linear 48B A3B InstructKimi | 2.46 | AA · 2,158 text-only questions · May 2025 revision |
| 598 | GPT-4o (Nov '24)OpenAI | 2.42 | AA · 2,158 text-only questions · May 2025 revision |
| 599 | GPT-4o (Aug '24)OpenAI | 2.33 | AA · 2,158 text-only questions · May 2025 revision |
| 600 | Qwen3 8B (Non-reasoning)Alibaba | 1.95 | AA · 2,158 text-only questions · May 2025 revision |
| 601 | GPT-4o (May '24)OpenAI | 1.76 | AA · 2,158 text-only questions · May 2025 revision |
| 602 | Qwen3.5 0.8B (Reasoning)Alibaba | 1.07 | AA · 2,158 text-only questions · May 2025 revision |
일치하는 모델이 없습니다. 검색어를 줄여 보세요.
공개된 숫자만 높은 순으로 표시합니다. 순서가 신뢰구간을 고려한 확정 순위는 아닙니다. 텍스트 2,158문항에 대한 AA의 평가입니다. 이미지 포함 전체 HLE나 도구 사용 결과와 구분합니다. 원문 모델명에 담긴 추론 강도와 fallback 설정을 그대로 보존합니다.
평가 방법 확인 ↗읽는 법
전체 벤치마크기존 benchmark가 쉬워진 뒤 나온 고난도 종합 시험형 benchmark입니다.
점수가 낮아도 이상한 게 아닙니다. 매우 어려운 문제에서 모델 간 차이를 보기 위한 용도입니다.