벤치마크 상세
GPQA Diamond
대학원 수준 과학 추론: 과학, 수학, 전문 지식 질문, 깊은 추론
공식 리더보드 보기
Artificial Analysis가 측정한 Diamond 결과만 사용합니다. 다른 회사의 자체 발표 점수는 섞지 않습니다. GPQA는 v4.2 종합 지수에서는 제외됐지만 개별 결과는 공개돼 있습니다.
모든 공개 결과
모델 검색 · 점수순확인 2026. 09. 07.Artificial Analysis · GPQA Diamond% correct
611개 결과원문 ↗
| 순서 | 모델 / 실행 구성 | 점수 | 평가 조건·근거 |
|---|---|---|---|
| 1 | GPT-6 Astra (xhigh)OpenAI | 96.26 | Artificial Analysis · GPQA Diamond |
| 2 | GPT-6 Astra (max)OpenAI | 96.06 | Artificial Analysis · GPQA Diamond |
| 3 | Gemini 3.8 Flash (high)Google | 95.25 | Artificial Analysis · GPQA Diamond |
| 4 | Grok 4.6 (high)SpaceXAI | 94.95 | Artificial Analysis · GPQA Diamond |
| 5 | GPT-6 Astra (high)OpenAI | 94.95 | Artificial Analysis · GPQA Diamond |
| 6 | Gemini 3.7 Flash (high)Google | 94.55 | Artificial Analysis · GPQA Diamond |
| 7 | Gemini 3.1 Pro PreviewGoogle | 94.14 | Artificial Analysis · GPQA Diamond |
| 8 | Muse Spark 1.3 (xhigh)Meta | 94.14 | Artificial Analysis · GPQA Diamond |
| 9 | GPT-5.6 Sol (max)OpenAI | 94.14 | Artificial Analysis · GPQA Diamond |
| 10 | GPT-6 Astra (medium)OpenAI | 93.94 | Artificial Analysis · GPQA Diamond |
| 11 | Muse Spark 1.3 (max)Meta | 93.84 | Artificial Analysis · GPQA Diamond |
| 12 | Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)Anthropic | 93.74 | Artificial Analysis · GPQA Diamond |
| 13 | Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic | 93.74 | Artificial Analysis · GPQA Diamond |
| 14 | Claude Opus 5 (Adaptive Reasoning, High Effort)Anthropic | 93.74 | Artificial Analysis · GPQA Diamond |
| 15 | GPT-5.5 (xhigh)OpenAI | 93.54 | Artificial Analysis · GPQA Diamond |
| 16 | Grok 4.6 (medium)SpaceXAI | 93.54 | Artificial Analysis · GPQA Diamond |
| 17 | Gemini 3.8 Flash (medium)Google | 93.54 | Artificial Analysis · GPQA Diamond |
| 18 | Qwen3.8 2.4T A95BAlibaba | 93.54 | Artificial Analysis · GPQA Diamond |
| 19 | Grok 4.6 (xhigh)SpaceXAI | 93.54 | Artificial Analysis · GPQA Diamond |
| 20 | Kimi K3 (max)Kimi | 93.54 | Artificial Analysis · GPQA Diamond |
| 21 | Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)Anthropic | 93.43 | Artificial Analysis · GPQA Diamond |
| 22 | GPT-5.5 (high)OpenAI | 93.23 | Artificial Analysis · GPQA Diamond |
| 23 | Claude Opus 5 (Adaptive Reasoning, Max Effort)Anthropic | 93.23 | Artificial Analysis · GPQA Diamond |
| 24 | Grok 4.5 (high)SpaceXAI | 93.13 | Artificial Analysis · GPQA Diamond |
| 25 | GPT-6 Astra (low)OpenAI | 93.13 | Artificial Analysis · GPQA Diamond |
| 26 | GPT-5.6 Sol (xhigh)OpenAI | 93.13 | Artificial Analysis · GPQA Diamond |
| 27 | MiniMax-M3MiniMax | 92.93 | Artificial Analysis · GPQA Diamond |
| 28 | Gemini 3.6 Flash (high)Google | 92.83 | Artificial Analysis · GPQA Diamond |
| 29 | DeepSeek V4 Pro 0813 (Reasoning, Max Effort)DeepSeek | 92.83 | Artificial Analysis · GPQA Diamond |
| 30 | GPT-5.6 Sol (high)OpenAI | 92.83 | Artificial Analysis · GPQA Diamond |
| 31 | Qwen3.8 MaxAlibaba | 92.73 | Artificial Analysis · GPQA Diamond |
| 32 | GPT-5.5 (medium)OpenAI | 92.63 | Artificial Analysis · GPQA Diamond |
| 33 | GPT-5.6 Sol (medium)OpenAI | 92.63 | Artificial Analysis · GPQA Diamond |
| 34 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)Anthropic | 92.63 | Artificial Analysis · GPQA Diamond |
| 35 | GPT-5.6 Terra (max)OpenAI | 92.53 | Artificial Analysis · GPQA Diamond |
| 36 | Qwen3.7 MaxAlibaba | 92.32 | Artificial Analysis · GPQA Diamond |
| 37 | Qwen3.8-Flash-NextAlibaba | 92.32 | Artificial Analysis · GPQA Diamond |
| 38 | Gemini 3.5 Flash (high)Google | 92.22 | Artificial Analysis · GPQA Diamond |
| 39 | Gemini 3.5 Flash (medium)Google | 92.12 | Artificial Analysis · GPQA Diamond |
| 40 | Gemini 3.7 Flash (medium)Google | 92.12 | Artificial Analysis · GPQA Diamond |
| 41 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic | 92.02 | Artificial Analysis · GPQA Diamond |
| 42 | GPT-5.4 (xhigh)OpenAI | 92.02 | Artificial Analysis · GPQA Diamond |
| 43 | Gemini 3.8 Flash (low)Google | 92.02 | Artificial Analysis · GPQA Diamond |
| 44 | Claude Opus 5 (Adaptive Reasoning, Medium Effort)Anthropic | 91.92 | Artificial Analysis · GPQA Diamond |
| 45 | GLM-5.3 (max)Z AI | 91.72 | Artificial Analysis · GPQA Diamond |
| 46 | GPT-5.3 Codex (xhigh)OpenAI | 91.52 | Artificial Analysis · GPQA Diamond |
| 47 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic | 91.41 | Artificial Analysis · GPQA Diamond |
| 48 | DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek | 91.31 | Artificial Analysis · GPQA Diamond |
| 49 | GLM-5.3-FlashZ AI | 91.21 | Artificial Analysis · GPQA Diamond |
| 50 | Kimi K2.6Kimi | 91.11 | Artificial Analysis · GPQA Diamond |
| 51 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort)Anthropic | 91.11 | Artificial Analysis · GPQA Diamond |
| 52 | GPT-5.6 Luna (max)OpenAI | 91.11 | Artificial Analysis · GPQA Diamond |
| 53 | Grok 4.20 0309 v2 (Reasoning)SpaceXAI | 91.11 | Artificial Analysis · GPQA Diamond |
| 54 | GPT-5.5 (low)OpenAI | 91.01 | Artificial Analysis · GPQA Diamond |
| 55 | GPT-5.6 Terra (xhigh)OpenAI | 90.81 | Artificial Analysis · GPQA Diamond |
| 56 | Gemini 3 Pro Preview (high)Google | 90.81 | Artificial Analysis · GPQA Diamond |
| 57 | DeepSeek V4 Flash 0731 (Reasoning, Max Effort)DeepSeek | 90.81 | Artificial Analysis · GPQA Diamond |
| 58 | Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)Anthropic | 90.61 | Artificial Analysis · GPQA Diamond |
| 59 | DeepSeek V4 Pro (Reasoning, High Effort)DeepSeek | 90.51 | Artificial Analysis · GPQA Diamond |
| 60 | Qwen3.8 27B (xhigh)Alibaba | 90.51 | Artificial Analysis · GPQA Diamond |
| 61 | Agnes 2.5 Pro BetaSapiens AI | 90.51 | Artificial Analysis · GPQA Diamond |
| 62 | Muse Spark 1.2 (xhigh)Meta | 90.4 | Artificial Analysis · GPQA Diamond |
| 63 | GPT-5.2 (xhigh)OpenAI | 90.3 | Artificial Analysis · GPQA Diamond |
| 64 | Grok 4.3 (high)SpaceXAI | 90.1 | Artificial Analysis · GPQA Diamond |
| 65 | Gemini 3.7 Flash (low)Google | 90.1 | Artificial Analysis · GPQA Diamond |
| 66 | Qwen3.7 PlusAlibaba | 90 | Artificial Analysis · GPQA Diamond |
| 67 | GPT-5.2 Codex (xhigh)OpenAI | 89.9 | Artificial Analysis · GPQA Diamond |
| 68 | GPT-5.6 Sol (low)OpenAI | 89.8 | Artificial Analysis · GPQA Diamond |
| 69 | Gemini 3 Flash Preview (Reasoning)Google | 89.8 | Artificial Analysis · GPQA Diamond |
| 70 | Muse Spark 1.1 (xhigh)Meta | 89.8 | Artificial Analysis · GPQA Diamond |
| 71 | Hy3Tencent | 89.7 | Artificial Analysis · GPQA Diamond |
| 72 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort)Anthropic | 89.6 | Artificial Analysis · GPQA Diamond |
| 73 | GPT-5.6 Terra (high)OpenAI | 89.6 | Artificial Analysis · GPQA Diamond |
| 74 | Kimi K2.7 CodeKimi | 89.6 | Artificial Analysis · GPQA Diamond |
| 75 | Grok Build 0.1 0616SpaceXAI | 89.49 | Artificial Analysis · GPQA Diamond |
| 76 | GPT-6 Astra (Non-reasoning)OpenAI | 89.49 | Artificial Analysis · GPQA Diamond |
| 77 | Inkling SmallThinking Machines | 89.49 | Artificial Analysis · GPQA Diamond |
| 78 | GPT-5.6 Luna (xhigh)OpenAI | 89.49 | Artificial Analysis · GPQA Diamond |
| 79 | GLM-5.2 (max)Z AI | 89.49 | Artificial Analysis · GPQA Diamond |
| 80 | DeepSeek V4 Flash (Reasoning, Max Effort)DeepSeek | 89.39 | Artificial Analysis · GPQA Diamond |
| 81 | Qwen3.5 397B A17B (Reasoning)Alibaba | 89.29 | Artificial Analysis · GPQA Diamond |
| 82 | Nex-N2-ProNex AGI | 89.19 | Artificial Analysis · GPQA Diamond |
| 83 | GPT-5.6 Luna (high)OpenAI | 89.19 | Artificial Analysis · GPQA Diamond |
| 84 | Solar Pro 4Upstage | 89.09 | Artificial Analysis · GPQA Diamond |
| 85 | Grok 4.3 (medium)SpaceXAI | 88.99 | Artificial Analysis · GPQA Diamond |
| 86 | Claude Opus 5 (Adaptive Reasoning, Low Effort)Anthropic | 88.89 | Artificial Analysis · GPQA Diamond |
| 87 | Qwen3.6 Max PreviewAlibaba | 88.79 | Artificial Analysis · GPQA Diamond |
| 88 | DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek | 88.79 | Artificial Analysis · GPQA Diamond |
| 89 | Gemini 3 Pro Preview (low)Google | 88.69 | Artificial Analysis · GPQA Diamond |
| 90 | Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback)Anthropic | 88.59 | Artificial Analysis · GPQA Diamond |
| 91 | Grok 4.20 0309 (Reasoning)SpaceXAI | 88.48 | Artificial Analysis · GPQA Diamond |
| 92 | Claude Opus 4.7 (Non-reasoning, High Effort)Anthropic | 88.48 | Artificial Analysis · GPQA Diamond |
| 93 | Muse SparkMeta | 88.38 | Artificial Analysis · GPQA Diamond |
| 94 | Qwen3.6 PlusAlibaba | 88.18 | Artificial Analysis · GPQA Diamond |
| 95 | Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback)Anthropic | 88.08 | Artificial Analysis · GPQA Diamond |
| 96 | Kimi K2.5 (Reasoning)Kimi | 87.88 | Artificial Analysis · GPQA Diamond |
| 97 | Grok 4.6 (low)SpaceXAI | 87.88 | Artificial Analysis · GPQA Diamond |
| 98 | Grok 4SpaceXAI | 87.68 | Artificial Analysis · GPQA Diamond |
| 99 | Agnes 2.5 Pro AlphaSapiens AI | 87.58 | Artificial Analysis · GPQA Diamond |
| 100 | GPT-5.4 mini (xhigh)OpenAI | 87.47 | Artificial Analysis · GPQA Diamond |
| 101 | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)Anthropic | 87.47 | Artificial Analysis · GPQA Diamond |
| 102 | MiniMax-M2.7MiniMax | 87.37 | Artificial Analysis · GPQA Diamond |
| 103 | GPT-5.1 (high)OpenAI | 87.27 | Artificial Analysis · GPQA Diamond |
| 104 | K2 Horizon 375B A23BMBZUAI Institute of Foundation Models | 87.27 | Artificial Analysis · GPQA Diamond |
| 105 | Inkling (xhigh)Thinking Machines | 87.17 | Artificial Analysis · GPQA Diamond |
| 106 | GPT-5.6 Terra (medium)OpenAI | 87.17 | Artificial Analysis · GPQA Diamond |
| 107 | DeepSeek V3.2 SpecialeDeepSeek | 87.07 | Artificial Analysis · GPQA Diamond |
| 108 | GPT-5.4 (low)OpenAI | 87.07 | Artificial Analysis · GPQA Diamond |
| 109 | MiMo-V2-ProXiaomi | 86.97 | Artificial Analysis · GPQA Diamond |
| 110 | Motif 3 (Beta)Motif Technologies | 86.87 | Artificial Analysis · GPQA Diamond |
| 111 | GLM-5.1 (Reasoning)Z AI | 86.77 | Artificial Analysis · GPQA Diamond |
| 112 | Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA | 86.67 | Artificial Analysis · GPQA Diamond |
| 113 | DeepSeek V4 Flash (Reasoning, High Effort)DeepSeek | 86.67 | Artificial Analysis · GPQA Diamond |
| 114 | Hy3-preview (Reasoning)Tencent | 86.67 | Artificial Analysis · GPQA Diamond |
| 115 | MiMo-V2.5-ProXiaomi | 86.57 | Artificial Analysis · GPQA Diamond |
| 116 | Claude Opus 4.5 (Reasoning)Anthropic | 86.57 | Artificial Analysis · GPQA Diamond |
| 117 | Apodex 1.1Apodex | 86.36 | Artificial Analysis · GPQA Diamond |
| 118 | GPT-5.2 (medium)OpenAI | 86.36 | Artificial Analysis · GPQA Diamond |
| 119 | Qwen3.5 397B A17B (Non-reasoning)Alibaba | 86.06 | Artificial Analysis · GPQA Diamond |
| 120 | Qwen3 Max ThinkingAlibaba | 86.06 | Artificial Analysis · GPQA Diamond |
| 121 | GPT-5.1 Codex (high)OpenAI | 85.96 | Artificial Analysis · GPQA Diamond |
| 122 | GPT-5.6 Luna (medium)OpenAI | 85.86 | Artificial Analysis · GPQA Diamond |
| 123 | GLM-4.7 (Reasoning)Z AI | 85.86 | Artificial Analysis · GPQA Diamond |
| 124 | Qwen3.5 27B (Reasoning)Alibaba | 85.76 | Artificial Analysis · GPQA Diamond |
| 125 | Qwen3.5 122B A10B (Reasoning)Alibaba | 85.66 | Artificial Analysis · GPQA Diamond |
| 126 | Solar Open2 250BUpstage | 85.66 | Artificial Analysis · GPQA Diamond |
| 127 | A.X-K2SK Telecom | 85.66 | Artificial Analysis · GPQA Diamond |
| 128 | Gemma 4 31B (Reasoning)Google | 85.66 | Artificial Analysis · GPQA Diamond |
| 129 | Ring-2.6-1TInclusionAI | 85.66 | Artificial Analysis · GPQA Diamond |
| 130 | MiMo-V2-Omni-0327Xiaomi | 85.45 | Artificial Analysis · GPQA Diamond |
| 131 | Ling 3.0 FlashInclusionAI | 85.45 | Artificial Analysis · GPQA Diamond |
| 132 | KAT Coder Pro V2KwaiKAT | 85.45 | Artificial Analysis · GPQA Diamond |
| 133 | GPT-5 (high)OpenAI | 85.35 | Artificial Analysis · GPQA Diamond |
| 134 | Grok 4.1 Fast (Reasoning)SpaceXAI | 85.25 | Artificial Analysis · GPQA Diamond |
| 135 | MiMo-V2.5Xiaomi | 84.95 | Artificial Analysis · GPQA Diamond |
| 136 | Nanbeige4.1-3BNanbeige | 84.95 | Artificial Analysis · GPQA Diamond |
| 137 | MiniMax-M2.5MiniMax | 84.85 | Artificial Analysis · GPQA Diamond |
| 138 | GLM-5-TurboZ AI | 84.75 | Artificial Analysis · GPQA Diamond |
| 139 | Grok 4 Fast (Reasoning)SpaceXAI | 84.75 | Artificial Analysis · GPQA Diamond |
| 140 | MiMo-V2-Flash (Reasoning)Xiaomi | 84.65 | Artificial Analysis · GPQA Diamond |
| 141 | GPT-5.5 Instant (May 2026)OpenAI | 84.65 | Artificial Analysis · GPQA Diamond |
| 142 | Qwen3.5 35B A3B (Reasoning)Alibaba | 84.55 | Artificial Analysis · GPQA Diamond |
| 143 | Qwen3.8 27B (medium)Alibaba | 84.55 | Artificial Analysis · GPQA Diamond |
| 144 | Qwen3.8 27B (low)Alibaba | 84.55 | Artificial Analysis · GPQA Diamond |
| 145 | JT-4.1 Flash 236B A21BChina Mobile | 84.55 | Artificial Analysis · GPQA Diamond |
| 146 | o3-proOpenAI | 84.55 | Artificial Analysis · GPQA Diamond |
| 147 | Gemini 2.5 ProGoogle | 84.44 | Artificial Analysis · GPQA Diamond |
| 148 | GPT-5.6 Terra (low)OpenAI | 84.34 | Artificial Analysis · GPQA Diamond |
| 149 | Grok 4.3 (low)SpaceXAI | 84.34 | Artificial Analysis · GPQA Diamond |
| 150 | Qwen3.5 27B (Non-reasoning)Alibaba | 84.24 | Artificial Analysis · GPQA Diamond |
| 151 | Kimi K3 (low)Kimi | 84.24 | Artificial Analysis · GPQA Diamond |
| 152 | Qwen3.6 27B (Reasoning)Alibaba | 84.24 | Artificial Analysis · GPQA Diamond |
| 153 | GPT-5 (medium)OpenAI | 84.18 | Artificial Analysis · GPQA Diamond |
| 154 | Qwen3.6 35B A3B (Reasoning)Alibaba | 84.14 | Artificial Analysis · GPQA Diamond |
| 155 | Claude Opus 4.6 (Non-reasoning, High Effort)Anthropic | 84.04 | Artificial Analysis · GPQA Diamond |
| 156 | DeepSeek V3.2 (Reasoning)DeepSeek | 84.04 | Artificial Analysis · GPQA Diamond |
| 157 | GLM-5.1 (Non-reasoning)Z AI | 83.94 | Artificial Analysis · GPQA Diamond |
| 158 | Gemini 3.5 Flash-LiteGoogle | 83.84 | Artificial Analysis · GPQA Diamond |
| 159 | Kimi K2 ThinkingKimi | 83.84 | Artificial Analysis · GPQA Diamond |
| 160 | GPT-5 Codex (high)OpenAI | 83.74 | Artificial Analysis · GPQA Diamond |
| 161 | Gemini 2.5 Pro Preview (Mar' 25)Google | 83.64 | Artificial Analysis · GPQA Diamond |
| 162 | GPT-5.6 Luna (low)OpenAI | 83.54 | Artificial Analysis · GPQA Diamond |
| 163 | MiMo-V2-Flash (Feb 2026)Xiaomi | 83.54 | Artificial Analysis · GPQA Diamond |
| 164 | Muse Glimmer (high)Meta | 83.54 | Artificial Analysis · GPQA Diamond |
| 165 | Claude 4.5 Sonnet (Reasoning)Anthropic | 83.43 | Artificial Analysis · GPQA Diamond |
| 166 | Motif 3Motif Technologies | 83.43 | Artificial Analysis · GPQA Diamond |
| 167 | Step 3.5 FlashStepFun | 83.13 | Artificial Analysis · GPQA Diamond |
| 168 | MiniMax-M2.1MiniMax | 83.03 | Artificial Analysis · GPQA Diamond |
| 169 | JT-35B-FlashChina Mobile | 82.93 | Artificial Analysis · GPQA Diamond |
| 170 | K-EXAONE 2.0LG AI Research | 82.93 | Artificial Analysis · GPQA Diamond |
| 171 | Qwen3.6 27B (Non-reasoning)Alibaba | 82.93 | Artificial Analysis · GPQA Diamond |
| 172 | Gemini 3.5 Flash (minimal)Google | 82.83 | Artificial Analysis · GPQA Diamond |
| 173 | GPT-5 mini (high)OpenAI | 82.83 | Artificial Analysis · GPQA Diamond |
| 174 | MiMo-V2-OmniXiaomi | 82.83 | Artificial Analysis · GPQA Diamond |
| 175 | Qwen3.5 122B A10B (Non-reasoning)Alibaba | 82.73 | Artificial Analysis · GPQA Diamond |
| 176 | o3OpenAI | 82.73 | Artificial Analysis · GPQA Diamond |
| 177 | Qwen3.5 Omni PlusAlibaba | 82.63 | Artificial Analysis · GPQA Diamond |
| 178 | Step 3.5 Flash 2603StepFun | 82.63 | Artificial Analysis · GPQA Diamond |
| 179 | GPT-5.4 mini (medium)OpenAI | 82.32 | Artificial Analysis · GPQA Diamond |
| 180 | GPT-5.5 Instant (June 2026)OpenAI | 82.32 | Artificial Analysis · GPQA Diamond |
| 181 | Gemini 2.5 Pro Preview (May' 25)Google | 82.22 | Artificial Analysis · GPQA Diamond |
| 182 | Gemini 3.1 Flash-LiteGoogle | 82.22 | Artificial Analysis · GPQA Diamond |
| 183 | GLM-5 (Reasoning)Z AI | 82.02 | Artificial Analysis · GPQA Diamond |
| 184 | Qwen3.5 35B A3B (Non-reasoning)Alibaba | 81.92 | Artificial Analysis · GPQA Diamond |
| 185 | Qwen3.8 27B (Non-reasoning)Alibaba | 81.82 | Artificial Analysis · GPQA Diamond |
| 186 | Qwen3.6 35B A3B (Non-reasoning)Alibaba | 81.72 | Artificial Analysis · GPQA Diamond |
| 187 | GPT-5.4 nano (xhigh)OpenAI | 81.72 | Artificial Analysis · GPQA Diamond |
| 188 | DeepSeek R1 0528 (May '25)DeepSeek | 81.31 | Artificial Analysis · GPQA Diamond |
| 189 | GPT-5.1 Codex mini (high)OpenAI | 81.31 | Artificial Analysis · GPQA Diamond |
| 190 | Gemini 3 Flash Preview (Non-reasoning)Google | 81.21 | Artificial Analysis · GPQA Diamond |
| 191 | Nova 2.0 Lite (high)Amazon | 81.11 | Artificial Analysis · GPQA Diamond |
| 192 | ERNIE 4.5 300B A47BBaidu | 81.11 | Artificial Analysis · GPQA Diamond |
| 193 | Claude Opus 4.5 (Non-reasoning)Anthropic | 81.01 | Artificial Analysis · GPQA Diamond |
| 194 | GLM 5V Turbo (Reasoning)Z AI | 80.91 | Artificial Analysis · GPQA Diamond |
| 195 | Step 3.7 FlashStepFun | 80.91 | Artificial Analysis · GPQA Diamond |
| 196 | Claude 4.1 Opus (Reasoning)Anthropic | 80.9 | Artificial Analysis · GPQA Diamond |
| 197 | GPT-5 (low)OpenAI | 80.81 | Artificial Analysis · GPQA Diamond |
| 198 | Qwen3.5 9B (Reasoning)Alibaba | 80.61 | Artificial Analysis · GPQA Diamond |
| 199 | G9v3-39A5BAI9Stars | 80.51 | Artificial Analysis · GPQA Diamond |
| 200 | GPT-5 mini (medium)OpenAI | 80.3 | Artificial Analysis · GPQA Diamond |
| 201 | Nemotron 3 Super 120B A12B (Reasoning)NVIDIA | 80 | Artificial Analysis · GPQA Diamond |
| 202 | Claude Sonnet 5 (Non-reasoning, High Effort)Anthropic | 80 | Artificial Analysis · GPQA Diamond |
| 203 | Claude Sonnet 4.6 (Non-reasoning, High Effort)Anthropic | 79.9 | Artificial Analysis · GPQA Diamond |
| 204 | DeepSeek V3.2 Exp (Reasoning)DeepSeek | 79.7 | Artificial Analysis · GPQA Diamond |
| 205 | Claude Sonnet 4.6 (Non-reasoning, Low Effort)Anthropic | 79.7 | Artificial Analysis · GPQA Diamond |
| 206 | Claude 4 Opus (Reasoning)Anthropic | 79.6 | Artificial Analysis · GPQA Diamond |
| 207 | EXAONE 4.5 33BLG AI Research | 79.39 | Artificial Analysis · GPQA Diamond |
| 208 | Gemini 2.5 Flash Preview (Sep '25) (Reasoning)Google | 79.29 | Artificial Analysis · GPQA Diamond |
| 209 | DeepSeek V3.1 Terminus (Reasoning)DeepSeek | 79.19 | Artificial Analysis · GPQA Diamond |
| 210 | Gemma 4 26B A4B (Reasoning)Google | 79.19 | Artificial Analysis · GPQA Diamond |
| 211 | Grok 3 mini Reasoning (high)SpaceXAI | 79.09 | Artificial Analysis · GPQA Diamond |
| 212 | Gemini 2.5 Flash (Reasoning)Google | 78.99 | Artificial Analysis · GPQA Diamond |
| 213 | GPT-5.6 Sol (Non-reasoning)OpenAI | 78.99 | Artificial Analysis · GPQA Diamond |
| 214 | Qwen3 235B A22B 2507 (Reasoning)Alibaba | 78.99 | Artificial Analysis · GPQA Diamond |
| 215 | Kimi K2.5 (Non-reasoning)Kimi | 78.89 | Artificial Analysis · GPQA Diamond |
| 216 | Kimi K2.6 (Non-reasoning)Kimi | 78.79 | Artificial Analysis · GPQA Diamond |
| 217 | Qwen3.5 9B (Non-reasoning)Alibaba | 78.59 | Artificial Analysis · GPQA Diamond |
| 218 | Nova 2.0 Pro Preview (medium)Amazon | 78.48 | Artificial Analysis · GPQA Diamond |
| 219 | Grok 4.20 0309 (Non-reasoning)SpaceXAI | 78.48 | Artificial Analysis · GPQA Diamond |
| 220 | o4-mini (high)OpenAI | 78.38 | Artificial Analysis · GPQA Diamond |
| 221 | K-EXAONE (Reasoning)LG AI Research | 78.28 | Artificial Analysis · GPQA Diamond |
| 222 | GLM-4.5 (Reasoning)Z AI | 78.18 | Artificial Analysis · GPQA Diamond |
| 223 | gpt-oss-120b (high)OpenAI | 78.18 | Artificial Analysis · GPQA Diamond |
| 224 | LongCat 2.0LongCat | 77.98 | Artificial Analysis · GPQA Diamond |
| 225 | GLM-4.6 (Reasoning)Z AI | 77.98 | Artificial Analysis · GPQA Diamond |
| 226 | DeepSeek V3.1 (Reasoning)DeepSeek | 77.88 | Artificial Analysis · GPQA Diamond |
| 227 | ERNIE 5.0 Thinking PreviewBaidu | 77.68 | Artificial Analysis · GPQA Diamond |
| 228 | Claude 4 Sonnet (Reasoning)Anthropic | 77.68 | Artificial Analysis · GPQA Diamond |
| 229 | MiniMax-M2MiniMax | 77.68 | Artificial Analysis · GPQA Diamond |
| 230 | Grok 4.20 0309 v2 (Non-reasoning)SpaceXAI | 77.58 | Artificial Analysis · GPQA Diamond |
| 231 | Qwen3 Max Thinking (Preview)Alibaba | 77.58 | Artificial Analysis · GPQA Diamond |
| 232 | Ring-1TInclusionAI | 77.37 | Artificial Analysis · GPQA Diamond |
| 233 | o3-mini (high)OpenAI | 77.27 | Artificial Analysis · GPQA Diamond |
| 234 | Claude 3.7 Sonnet (Reasoning)Anthropic | 77.17 | Artificial Analysis · GPQA Diamond |
| 235 | Qwen3 VL 235B A22B (Reasoning)Alibaba | 77.17 | Artificial Analysis · GPQA Diamond |
| 236 | Qwen3.5 4B (Reasoning)Alibaba | 77.07 | Artificial Analysis · GPQA Diamond |
| 237 | Mercury 2Inception | 76.97 | Artificial Analysis · GPQA Diamond |
| 238 | Mistral Small 4 (Reasoning)Mistral | 76.87 | Artificial Analysis · GPQA Diamond |
| 239 | GPT-5.5 (Non-reasoning)OpenAI | 76.77 | Artificial Analysis · GPQA Diamond |
| 240 | Cogito v2.1 (Reasoning)Deep Cogito | 76.77 | Artificial Analysis · GPQA Diamond |
| 241 | Nova 2.0 Lite (medium)Amazon | 76.77 | Artificial Analysis · GPQA Diamond |
| 242 | Kimi K2 0905Kimi | 76.67 | Artificial Analysis · GPQA Diamond |
| 243 | Kimi K2Kimi | 76.57 | Artificial Analysis · GPQA Diamond |
| 244 | Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)Google | 76.57 | Artificial Analysis · GPQA Diamond |
| 245 | o1-previewOpenAI | 76.46 | Artificial Analysis · GPQA Diamond |
| 246 | Doubao Seed CodeByteDance Seed | 76.36 | Artificial Analysis · GPQA Diamond |
| 247 | Qwen3 Max (Preview)Alibaba | 76.36 | Artificial Analysis · GPQA Diamond |
| 248 | Qwen3 MaxAlibaba | 76.36 | Artificial Analysis · GPQA Diamond |
| 249 | KAT-Coder-Pro V1KwaiKAT | 76.36 | Artificial Analysis · GPQA Diamond |
| 250 | Gemma 4 31B (Non-reasoning)Google | 76.26 | Artificial Analysis · GPQA Diamond |
| 251 | MiMo-V2.5-Pro (Non-reasoning)Xiaomi | 76.16 | Artificial Analysis · GPQA Diamond |
| 252 | Command A+Cohere | 76.06 | Artificial Analysis · GPQA Diamond |
| 253 | INTELLECT-3Prime Intellect | 76.06 | Artificial Analysis · GPQA Diamond |
| 254 | GPT-5.4 nano (medium)OpenAI | 76.06 | Artificial Analysis · GPQA Diamond |
| 255 | Nova 2.0 Omni (medium)Amazon | 75.96 | Artificial Analysis · GPQA Diamond |
| 256 | Qwen3 Next 80B A3B (Reasoning)Alibaba | 75.86 | Artificial Analysis · GPQA Diamond |
| 257 | Nemotron Cascade 2 30B A3BNVIDIA | 75.76 | Artificial Analysis · GPQA Diamond |
| 258 | North Mini CodeCohere | 75.66 | Artificial Analysis · GPQA Diamond |
| 259 | NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA | 75.66 | Artificial Analysis · GPQA Diamond |
| 260 | Qwen3 235B A22B 2507 InstructAlibaba | 75.25 | Artificial Analysis · GPQA Diamond |
| 261 | Gemma 4 12B (Reasoning)Google | 75.25 | Artificial Analysis · GPQA Diamond |
| 262 | Ling-2.6-1TInclusionAI | 75.15 | Artificial Analysis · GPQA Diamond |
| 263 | Trinity Large ThinkingArcee AI | 75.15 | Artificial Analysis · GPQA Diamond |
| 264 | DeepSeek V3.2 (Non-reasoning)DeepSeek | 75.05 | Artificial Analysis · GPQA Diamond |
| 265 | DeepSeek V3.1 Terminus (Non-reasoning)DeepSeek | 75.05 | Artificial Analysis · GPQA Diamond |
| 266 | Nova 2.0 Pro Preview (low)Amazon | 75.05 | Artificial Analysis · GPQA Diamond |
| 267 | o3-miniOpenAI | 74.85 | Artificial Analysis · GPQA Diamond |
| 268 | GPT-5.4 (Non-reasoning)OpenAI | 74.85 | Artificial Analysis · GPQA Diamond |
| 269 | Mistral Medium 3.5Mistral | 74.85 | Artificial Analysis · GPQA Diamond |
| 270 | Llama Nemotron Super 49B v1.5 (Reasoning)NVIDIA | 74.85 | Artificial Analysis · GPQA Diamond |
| 271 | o1OpenAI | 74.75 | Artificial Analysis · GPQA Diamond |
| 272 | GPT-5.6 Terra (Non-reasoning)OpenAI | 74.65 | Artificial Analysis · GPQA Diamond |
| 273 | Nemotron 3.5 LightningNVIDIA | 74.34 | Artificial Analysis · GPQA Diamond |
| 274 | Qwen3.5 Omni FlashAlibaba | 74.24 | Artificial Analysis · GPQA Diamond |
| 275 | EXAONE 4.0 32B (Reasoning)LG AI Research | 73.94 | Artificial Analysis · GPQA Diamond |
| 276 | Magistral Medium 1.2Mistral | 73.94 | Artificial Analysis · GPQA Diamond |
| 277 | DeepSeek V3.2 Exp (Non-reasoning)DeepSeek | 73.84 | Artificial Analysis · GPQA Diamond |
| 278 | Qwen3 Next 80B A3B InstructAlibaba | 73.84 | Artificial Analysis · GPQA Diamond |
| 279 | Sarvam 105B (high)Sarvam | 73.84 | Artificial Analysis · GPQA Diamond |
| 280 | Qwen3 Coder NextAlibaba | 73.74 | Artificial Analysis · GPQA Diamond |
| 281 | DeepSeek V3.1 (Non-reasoning)DeepSeek | 73.54 | Artificial Analysis · GPQA Diamond |
| 282 | Ling 3.0 TinyInclusionAI | 73.43 | Artificial Analysis · GPQA Diamond |
| 283 | HyperNova 60B 2605 (high, based on gpt-oss-120b)Multiverse Computing | 73.33 | Artificial Analysis · GPQA Diamond |
| 284 | GLM-4.5-AirZ AI | 73.33 | Artificial Analysis · GPQA Diamond |
| 285 | Apriel-v1.6-15B-ThinkerServiceNow | 73.33 | Artificial Analysis · GPQA Diamond |
| 286 | Qwen3 VL 32B (Reasoning)Alibaba | 73.33 | Artificial Analysis · GPQA Diamond |
| 287 | Hy3-preview (Non-reasoning)Tencent | 73.23 | Artificial Analysis · GPQA Diamond |
| 288 | Quasar 438B (max, based on GLM-5.2)Multiverse Computing | 73.23 | Artificial Analysis · GPQA Diamond |
| 289 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)NVIDIA | 72.83 | Artificial Analysis · GPQA Diamond |
| 290 | Hermes 4 - Llama-3.1 405B (Reasoning)Nous Research | 72.73 | Artificial Analysis · GPQA Diamond |
| 291 | Claude 4.5 Sonnet (Non-reasoning)Anthropic | 72.73 | Artificial Analysis · GPQA Diamond |
| 292 | Grok Code Fast 1SpaceXAI | 72.73 | Artificial Analysis · GPQA Diamond |
| 293 | Seed-OSS-36B-InstructByteDance Seed | 72.63 | Artificial Analysis · GPQA Diamond |
| 294 | Qwen3 Omni 30B A3B (Reasoning)Alibaba | 72.63 | Artificial Analysis · GPQA Diamond |
| 295 | Ring-flash-2.0InclusionAI | 72.53 | Artificial Analysis · GPQA Diamond |
| 296 | Solar Pro 3Upstage | 72.42 | Artificial Analysis · GPQA Diamond |
| 297 | Mi:dm K 2.5 Pro PreviewKorea Telecom | 72.22 | Artificial Analysis · GPQA Diamond |
| 298 | Qwen3 VL 30B A3B (Reasoning)Alibaba | 72.02 | Artificial Analysis · GPQA Diamond |
| 299 | Ling-1TInclusionAI | 71.92 | Artificial Analysis · GPQA Diamond |
| 300 | GLM-4.6V (Reasoning)Z AI | 71.92 | Artificial Analysis · GPQA Diamond |
| 301 | DeepSeek V4 Pro (Non-reasoning)DeepSeek | 71.72 | Artificial Analysis · GPQA Diamond |
| 302 | DeepSeek V4 Flash (Non-reasoning)DeepSeek | 71.62 | Artificial Analysis · GPQA Diamond |
| 303 | Gemma 4 26B A4B (Non-reasoning)Google | 71.41 | Artificial Analysis · GPQA Diamond |
| 304 | K2 Think V2MBZUAI Institute of Foundation Models | 71.31 | Artificial Analysis · GPQA Diamond |
| 305 | Apriel-v1.5-15B-ThinkerServiceNow | 71.31 | Artificial Analysis · GPQA Diamond |
| 306 | Qwen3.5 4B (Non-reasoning)Alibaba | 71.21 | Artificial Analysis · GPQA Diamond |
| 307 | GPT-5.2 (Non-reasoning)OpenAI | 71.21 | Artificial Analysis · GPQA Diamond |
| 308 | Qwen3 VL 235B A22B InstructAlibaba | 71.21 | Artificial Analysis · GPQA Diamond |
| 309 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)Google | 70.91 | Artificial Analysis · GPQA Diamond |
| 310 | DeepSeek R1 (Jan '25)DeepSeek | 70.81 | Artificial Analysis · GPQA Diamond |
| 311 | Qwen3 30B A3B 2507 (Reasoning)Alibaba | 70.71 | Artificial Analysis · GPQA Diamond |
| 312 | Gemini 2.0 Flash Thinking Experimental (Jan '25)Google | 70.1 | Artificial Analysis · GPQA Diamond |
| 313 | Claude 4 Opus (Non-reasoning)Anthropic | 70.1 | Artificial Analysis · GPQA Diamond |
| 314 | Mi:dm K 2.5 ProKorea Telecom | 70.1 | Artificial Analysis · GPQA Diamond |
| 315 | Qwen3 235B A22B (Reasoning)Alibaba | 70 | Artificial Analysis · GPQA Diamond |
| 316 | Hermes 4 - Llama-3.1 70B (Reasoning)Nous Research | 69.9 | Artificial Analysis · GPQA Diamond |
| 317 | Nova 2.0 Omni (low)Amazon | 69.9 | Artificial Analysis · GPQA Diamond |
| 318 | Gemini 2.5 Flash Preview (Reasoning)Google | 69.8 | Artificial Analysis · GPQA Diamond |
| 319 | Nova 2.0 Lite (low)Amazon | 69.8 | Artificial Analysis · GPQA Diamond |
| 320 | MiniMax M1 80kMiniMax | 69.7 | Artificial Analysis · GPQA Diamond |
| 321 | Qwen3 VL 30B A3B InstructAlibaba | 69.49 | Artificial Analysis · GPQA Diamond |
| 322 | Motif-2-12.7B-ReasoningMotif Technologies | 69.49 | Artificial Analysis · GPQA Diamond |
| 323 | K-EXAONE (Non-reasoning)LG AI Research | 69.49 | Artificial Analysis · GPQA Diamond |
| 324 | Grok 3SpaceXAI | 69.29 | Artificial Analysis · GPQA Diamond |
| 325 | Step3 VL 10BStepFun | 68.99 | Artificial Analysis · GPQA Diamond |
| 326 | gpt-oss-20b (high)OpenAI | 68.79 | Artificial Analysis · GPQA Diamond |
| 327 | GPT-5 mini (minimal)OpenAI | 68.69 | Artificial Analysis · GPQA Diamond |
| 328 | Solar Pro 2 (Reasoning)Upstage | 68.69 | Artificial Analysis · GPQA Diamond |
| 329 | GLM-5.2 (Non-reasoning)Z AI | 68.59 | Artificial Analysis · GPQA Diamond |
| 330 | GPT-5 (ChatGPT)OpenAI | 68.59 | Artificial Analysis · GPQA Diamond |
| 331 | GLM-4.5V (Reasoning)Z AI | 68.38 | Artificial Analysis · GPQA Diamond |
| 332 | Gemini 2.5 Flash (Non-reasoning)Google | 68.28 | Artificial Analysis · GPQA Diamond |
| 333 | Claude 4 Sonnet (Non-reasoning)Anthropic | 68.28 | Artificial Analysis · GPQA Diamond |
| 334 | MiniMax M1 40kMiniMax | 68.18 | Artificial Analysis · GPQA Diamond |
| 335 | K2-V2 (high)MBZUAI Institute of Foundation Models | 68.08 | Artificial Analysis · GPQA Diamond |
| 336 | Mistral Large 3Mistral | 67.98 | Artificial Analysis · GPQA Diamond |
| 337 | Magistral Medium 1Mistral | 67.88 | Artificial Analysis · GPQA Diamond |
| 338 | JT-MINIChina Mobile | 67.58 | Artificial Analysis · GPQA Diamond |
| 339 | GPT-5 nano (high)OpenAI | 67.58 | Artificial Analysis · GPQA Diamond |
| 340 | GPT-5 (minimal)OpenAI | 67.27 | Artificial Analysis · GPQA Diamond |
| 341 | Claude 4.5 Haiku (Reasoning)Anthropic | 67.17 | Artificial Analysis · GPQA Diamond |
| 342 | gpt-oss-120b (low)OpenAI | 67.17 | Artificial Analysis · GPQA Diamond |
| 343 | Qwen3 VL 32B InstructAlibaba | 67.07 | Artificial Analysis · GPQA Diamond |
| 344 | Llama 4 MaverickMeta | 67.07 | Artificial Analysis · GPQA Diamond |
| 345 | GPT-5 nano (medium)OpenAI | 66.97 | Artificial Analysis · GPQA Diamond |
| 346 | DiffusionGemma 26B A4BGoogle | 66.87 | Artificial Analysis · GPQA Diamond |
| 347 | Qwen3 32B (Reasoning)Alibaba | 66.77 | Artificial Analysis · GPQA Diamond |
| 348 | Qwen3 4B 2507 (Reasoning)Alibaba | 66.67 | Artificial Analysis · GPQA Diamond |
| 349 | GPT-4.1OpenAI | 66.57 | Artificial Analysis · GPQA Diamond |
| 350 | GLM-5 (Non-reasoning)Z AI | 66.57 | Artificial Analysis · GPQA Diamond |
| 351 | GLM-4.7 (Non-reasoning)Z AI | 66.36 | Artificial Analysis · GPQA Diamond |
| 352 | GPT-4.1 miniOpenAI | 66.36 | Artificial Analysis · GPQA Diamond |
| 353 | Magistral Small 1.2Mistral | 66.26 | Artificial Analysis · GPQA Diamond |
| 354 | Gemma 4 12B (Non-reasoning)Google | 66.06 | Artificial Analysis · GPQA Diamond |
| 355 | Falcon-H1R-7BTII UAE | 66.06 | Artificial Analysis · GPQA Diamond |
| 356 | Qwen3 30B A3B 2507 InstructAlibaba | 65.86 | Artificial Analysis · GPQA Diamond |
| 357 | Grok 4.3 (Non-reasoning)SpaceXAI | 65.76 | Artificial Analysis · GPQA Diamond |
| 358 | Solar Open 100B (Reasoning)Upstage | 65.66 | Artificial Analysis · GPQA Diamond |
| 359 | Ling-flash-2.0InclusionAI | 65.66 | Artificial Analysis · GPQA Diamond |
| 360 | MiMo-V2-Flash (Non-reasoning)Xiaomi | 65.56 | Artificial Analysis · GPQA Diamond |
| 361 | Claude 3.7 Sonnet (Non-reasoning)Anthropic | 65.56 | Artificial Analysis · GPQA Diamond |
| 362 | GPT-4o (March 2025, chatgpt-4o-latest)OpenAI | 65.45 | Artificial Analysis · GPQA Diamond |
| 363 | DeepSeek V3 0324DeepSeek | 65.45 | Artificial Analysis · GPQA Diamond |
| 364 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)Google | 65.05 | Artificial Analysis · GPQA Diamond |
| 365 | Claude 4.5 Haiku (Non-reasoning)Anthropic | 64.65 | Artificial Analysis · GPQA Diamond |
| 366 | GPT-5.6 Luna (Non-reasoning)OpenAI | 64.55 | Artificial Analysis · GPQA Diamond |
| 367 | Granite 4.2 30BIBM | 64.44 | Artificial Analysis · GPQA Diamond |
| 368 | Llama 3.3 Nemotron Super 49B v1 (Reasoning)NVIDIA | 64.34 | Artificial Analysis · GPQA Diamond |
| 369 | GPT-5.1 (Non-reasoning)OpenAI | 64.34 | Artificial Analysis · GPQA Diamond |
| 370 | Magistral Small 1Mistral | 64.14 | Artificial Analysis · GPQA Diamond |
| 371 | Grok 4.1 Fast (Non-reasoning)SpaceXAI | 63.74 | Artificial Analysis · GPQA Diamond |
| 372 | LongCat Flash LiteLongCat | 63.64 | Artificial Analysis · GPQA Diamond |
| 373 | Nova 2.0 Pro Preview (Non-reasoning)Amazon | 63.64 | Artificial Analysis · GPQA Diamond |
| 374 | Gemini 2.0 Flash (experimental)Google | 63.64 | Artificial Analysis · GPQA Diamond |
| 375 | Sarvam 30B (high)Sarvam | 63.33 | Artificial Analysis · GPQA Diamond |
| 376 | GLM-4.6 (Non-reasoning)Z AI | 63.23 | Artificial Analysis · GPQA Diamond |
| 377 | Celeris-1Celeris | 63.13 | Artificial Analysis · GPQA Diamond |
| 378 | Granite 4.2 8BIBM | 63.13 | Artificial Analysis · GPQA Diamond |
| 379 | EXAONE 4.0 32B (Non-reasoning)LG AI Research | 62.83 | Artificial Analysis · GPQA Diamond |
| 380 | Gemini 2.5 Flash-Lite (Reasoning)Google | 62.53 | Artificial Analysis · GPQA Diamond |
| 381 | Sonar ReasoningPerplexity | 62.32 | Artificial Analysis · GPQA Diamond |
| 382 | Gemini 2.0 Flash (Feb '25)Google | 62.32 | Artificial Analysis · GPQA Diamond |
| 383 | Gemini 2.0 Pro Experimental (Feb '25)Google | 62.22 | Artificial Analysis · GPQA Diamond |
| 384 | Qwen3 Omni 30B A3B InstructAlibaba | 62.02 | Artificial Analysis · GPQA Diamond |
| 385 | Qwen3 Coder 480B A35B InstructAlibaba | 61.82 | Artificial Analysis · GPQA Diamond |
| 386 | Qwen3 30B A3B (Reasoning)Alibaba | 61.62 | Artificial Analysis · GPQA Diamond |
| 387 | HyperCLOVA X SEED Think (32B)Naver | 61.52 | Artificial Analysis · GPQA Diamond |
| 388 | DeepSeek R1 Distill Qwen 32BDeepSeek | 61.52 | Artificial Analysis · GPQA Diamond |
| 389 | Qwen3 235B A22B (Non-reasoning)Alibaba | 61.31 | Artificial Analysis · GPQA Diamond |
| 390 | DeepSeek R1 0528 Qwen3 8BDeepSeek | 61.21 | Artificial Analysis · GPQA Diamond |
| 391 | gpt-oss-20b (low)OpenAI | 61.11 | Artificial Analysis · GPQA Diamond |
| 392 | Olmo 3 32B ThinkAllen Institute for AI | 61.01 | Artificial Analysis · GPQA Diamond |
| 393 | Grok 4 Fast (Non-reasoning)SpaceXAI | 60.61 | Artificial Analysis · GPQA Diamond |
| 394 | GPT-5.4 mini (Non-Reasoning)OpenAI | 60.61 | Artificial Analysis · GPQA Diamond |
| 395 | Qwen3 14B (Reasoning)Alibaba | 60.4 | Artificial Analysis · GPQA Diamond |
| 396 | Nova 2.0 Lite (Non-reasoning)Amazon | 60.3 | Artificial Analysis · GPQA Diamond |
| 397 | o1-miniOpenAI | 60.3 | Artificial Analysis · GPQA Diamond |
| 398 | Tri-21B-ThinkTrillion Labs | 60.1 | Artificial Analysis · GPQA Diamond |
| 399 | Claude 3.5 Sonnet (Oct '24)Anthropic | 59.9 | Artificial Analysis · GPQA Diamond |
| 400 | K2-V2 (medium)MBZUAI Institute of Foundation Models | 59.8 | Artificial Analysis · GPQA Diamond |
| 401 | Devstral 2Mistral | 59.39 | Artificial Analysis · GPQA Diamond |
| 402 | Gemini 2.5 Flash Preview (Non-reasoning)Google | 59.39 | Artificial Analysis · GPQA Diamond |
| 403 | Ling 2.6 FlashInclusionAI | 59.29 | Artificial Analysis · GPQA Diamond |
| 404 | QwQ 32BAlibaba | 59.29 | Artificial Analysis · GPQA Diamond |
| 405 | Olmo 3.1 32B ThinkAllen Institute for AI | 59.09 | Artificial Analysis · GPQA Diamond |
| 406 | Gemini 1.5 Pro (Sep '24)Google | 58.89 | Artificial Analysis · GPQA Diamond |
| 407 | Qwen3 8B (Reasoning)Alibaba | 58.89 | Artificial Analysis · GPQA Diamond |
| 408 | Mistral Medium 3.1Mistral | 58.79 | Artificial Analysis · GPQA Diamond |
| 409 | Llama 4 ScoutMeta | 58.69 | Artificial Analysis · GPQA Diamond |
| 410 | Qwen2.5 MaxAlibaba | 58.69 | Artificial Analysis · GPQA Diamond |
| 411 | GLM-4.7-Flash (Reasoning)Z AI | 58.08 | Artificial Analysis · GPQA Diamond |
| 412 | Qwen3 VL 8B (Reasoning)Alibaba | 57.88 | Artificial Analysis · GPQA Diamond |
| 413 | Sonar ProPerplexity | 57.78 | Artificial Analysis · GPQA Diamond |
| 414 | Mistral Medium 3Mistral | 57.78 | Artificial Analysis · GPQA Diamond |
| 415 | Solar Pro 2 (Preview) (Reasoning)Upstage | 57.78 | Artificial Analysis · GPQA Diamond |
| 416 | Gemma 4 E4B (Reasoning)Google | 57.58 | Artificial Analysis · GPQA Diamond |
| 417 | Phi-4Microsoft | 57.47 | Artificial Analysis · GPQA Diamond |
| 418 | GLM-4.5V (Non-reasoning)Z AI | 57.27 | Artificial Analysis · GPQA Diamond |
| 419 | NVIDIA Nemotron Nano 12B v2 VL (Reasoning)NVIDIA | 57.17 | Artificial Analysis · GPQA Diamond |
| 420 | Ministral 3 14BMistral | 57.17 | Artificial Analysis · GPQA Diamond |
| 421 | Mistral Small 4 (Non-reasoning)Mistral | 57.07 | Artificial Analysis · GPQA Diamond |
| 422 | NVIDIA Nemotron Nano 9B V2 (Reasoning)NVIDIA | 56.97 | Artificial Analysis · GPQA Diamond |
| 423 | Nova PremierAmazon | 56.87 | Artificial Analysis · GPQA Diamond |
| 424 | GLM-4.6V (Non-reasoning)Z AI | 56.57 | Artificial Analysis · GPQA Diamond |
| 425 | Ling-mini-2.0InclusionAI | 56.16 | Artificial Analysis · GPQA Diamond |
| 426 | Solar Pro 2 (Non-reasoning)Upstage | 56.06 | Artificial Analysis · GPQA Diamond |
| 427 | Claude 3.5 Sonnet (June '24)Anthropic | 55.96 | Artificial Analysis · GPQA Diamond |
| 428 | Granite 4.2 3BIBM | 55.86 | Artificial Analysis · GPQA Diamond |
| 429 | GPT-5.4 nano (Non-Reasoning)OpenAI | 55.76 | Artificial Analysis · GPQA Diamond |
| 430 | LFM2.5-2.6BLiquid AI | 55.76 | Artificial Analysis · GPQA Diamond |
| 431 | NVIDIA Nemotron Nano 9B V2 (Non-reasoning)NVIDIA | 55.66 | Artificial Analysis · GPQA Diamond |
| 432 | DeepSeek V3 (Dec '24)DeepSeek | 55.66 | Artificial Analysis · GPQA Diamond |
| 433 | QwQ 32B-PreviewAlibaba | 55.66 | Artificial Analysis · GPQA Diamond |
| 434 | Nova 2.0 Omni (Non-reasoning)Amazon | 55.45 | Artificial Analysis · GPQA Diamond |
| 435 | Gemma 4 E4B (Non-reasoning)Google | 54.95 | Artificial Analysis · GPQA Diamond |
| 436 | Solar Pro 2 (Preview) (Non-reasoning)Upstage | 54.44 | Artificial Analysis · GPQA Diamond |
| 437 | GPT-4o (Nov '24)OpenAI | 54.34 | Artificial Analysis · GPQA Diamond |
| 438 | Gemini 2.0 Flash-Lite (Preview)Google | 54.24 | Artificial Analysis · GPQA Diamond |
| 439 | K2-V2 (low)MBZUAI Institute of Foundation Models | 54.14 | Artificial Analysis · GPQA Diamond |
| 440 | Olmo 3.1 32B InstructAllen Institute for AI | 53.94 | Artificial Analysis · GPQA Diamond |
| 441 | Tri-21B-think PreviewTrillion Labs | 53.84 | Artificial Analysis · GPQA Diamond |
| 442 | Hermes 4 - Llama-3.1 405B (Non-reasoning)Nous Research | 53.64 | Artificial Analysis · GPQA Diamond |
| 443 | Gemini 2.0 Flash-Lite (Feb '25)Google | 53.54 | Artificial Analysis · GPQA Diamond |
| 444 | Qwen3 32B (Non-reasoning)Alibaba | 53.54 | Artificial Analysis · GPQA Diamond |
| 445 | Devstral Small 2Mistral | 53.23 | Artificial Analysis · GPQA Diamond |
| 446 | Reka Flash 3Reka AI | 52.93 | Artificial Analysis · GPQA Diamond |
| 447 | Command ACohere | 52.73 | Artificial Analysis · GPQA Diamond |
| 448 | GPT-4o (May '24)OpenAI | 52.63 | Artificial Analysis · GPQA Diamond |
| 449 | Qwen3 4B (Reasoning)Alibaba | 52.22 | Artificial Analysis · GPQA Diamond |
| 450 | GPT-4o (Aug '24)OpenAI | 52.12 | Artificial Analysis · GPQA Diamond |
| 451 | Qwen3 4B 2507 InstructAlibaba | 51.72 | Artificial Analysis · GPQA Diamond |
| 452 | Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)NVIDIA | 51.72 | Artificial Analysis · GPQA Diamond |
| 453 | Olmo 3 7B ThinkAllen Institute for AI | 51.62 | Artificial Analysis · GPQA Diamond |
| 454 | Qwen3 Coder 30B A3B InstructAlibaba | 51.62 | Artificial Analysis · GPQA Diamond |
| 455 | Llama 3.1 Tulu3 405BAllen Institute for AI | 51.62 | Artificial Analysis · GPQA Diamond |
| 456 | Llama 3.1 Instruct 405BMeta | 51.52 | Artificial Analysis · GPQA Diamond |
| 457 | Exaone 4.0 1.2B (Reasoning)LG AI Research | 51.52 | Artificial Analysis · GPQA Diamond |
| 458 | Qwen3 30B A3B (Non-reasoning)Alibaba | 51.52 | Artificial Analysis · GPQA Diamond |
| 459 | NVIDIA Nemotron 3 Nano 4BNVIDIA | 51.31 | Artificial Analysis · GPQA Diamond |
| 460 | LFM2.5-8B-A1BLiquid AI | 51.31 | Artificial Analysis · GPQA Diamond |
| 461 | GPT-4.1 nanoOpenAI | 51.21 | Artificial Analysis · GPQA Diamond |
| 462 | GPT-4o (ChatGPT)OpenAI | 51.11 | Artificial Analysis · GPQA Diamond |
| 463 | Grok 2 (Dec '24)SpaceXAI | 51.01 | Artificial Analysis · GPQA Diamond |
| 464 | Mistral Small 3.2Mistral | 50.51 | Artificial Analysis · GPQA Diamond |
| 465 | Pixtral LargeMistral | 50.51 | Artificial Analysis · GPQA Diamond |
| 466 | Nova ProAmazon | 49.9 | Artificial Analysis · GPQA Diamond |
| 467 | Llama 3.3 Instruct 70BMeta | 49.8 | Artificial Analysis · GPQA Diamond |
| 468 | Qwen3 VL 4B (Reasoning)Alibaba | 49.39 | Artificial Analysis · GPQA Diamond |
| 469 | Devstral MediumMistral | 49.19 | Artificial Analysis · GPQA Diamond |
| 470 | Hermes 4 - Llama-3.1 70B (Non-reasoning)Nous Research | 49.09 | Artificial Analysis · GPQA Diamond |
| 471 | Qwen2.5 Instruct 72BAlibaba | 49.09 | Artificial Analysis · GPQA Diamond |
| 472 | Claude 3 OpusAnthropic | 48.89 | Artificial Analysis · GPQA Diamond |
| 473 | Mistral Large 2 (Nov '24)Mistral | 48.59 | Artificial Analysis · GPQA Diamond |
| 474 | DeepSeek R1 Distill Qwen 14BDeepSeek | 48.38 | Artificial Analysis · GPQA Diamond |
| 475 | Llama Nemotron Super 49B v1.5 (Non-reasoning)NVIDIA | 48.08 | Artificial Analysis · GPQA Diamond |
| 476 | Granite 4.1 30BIBM | 48.08 | Artificial Analysis · GPQA Diamond |
| 477 | Gemini 2.5 Flash-Lite (Non-reasoning)Google | 47.37 | Artificial Analysis · GPQA Diamond |
| 478 | LFM2 24B A2BLiquid AI | 47.37 | Artificial Analysis · GPQA Diamond |
| 479 | Mistral Large 2 (Jul '24)Mistral | 47.17 | Artificial Analysis · GPQA Diamond |
| 480 | SonarPerplexity | 47.07 | Artificial Analysis · GPQA Diamond |
| 481 | Ministral 3 8BMistral | 47.07 | Artificial Analysis · GPQA Diamond |
| 482 | Grok BetaSpaceXAI | 47.07 | Artificial Analysis · GPQA Diamond |
| 483 | Qwen3 14B (Non-reasoning)Alibaba | 46.97 | Artificial Analysis · GPQA Diamond |
| 484 | Nemotron 3 Nano Omni 30B A3B ReasoningNVIDIA | 46.87 | Artificial Analysis · GPQA Diamond |
| 485 | Qwen2.5 Instruct 32BAlibaba | 46.57 | Artificial Analysis · GPQA Diamond |
| 486 | Llama 3.1 Nemotron Instruct 70BNVIDIA | 46.46 | Artificial Analysis · GPQA Diamond |
| 487 | Gemini 1.5 Flash (Sep '24)Google | 46.26 | Artificial Analysis · GPQA Diamond |
| 488 | Mistral Small 3Mistral | 46.16 | Artificial Analysis · GPQA Diamond |
| 489 | Qwen3.5 2B (Reasoning)Alibaba | 45.56 | Artificial Analysis · GPQA Diamond |
| 490 | Mistral Small 3.1Mistral | 45.35 | Artificial Analysis · GPQA Diamond |
| 491 | Qwen3 8B (Non-reasoning)Alibaba | 45.15 | Artificial Analysis · GPQA Diamond |
| 492 | GLM-4.7-Flash (Non-reasoning)Z AI | 45.15 | Artificial Analysis · GPQA Diamond |
| 493 | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)NVIDIA | 43.94 | Artificial Analysis · GPQA Diamond |
| 494 | Qwen3.5 2B (Non-reasoning)Alibaba | 43.84 | Artificial Analysis · GPQA Diamond |
| 495 | G9v3-3BAI9Stars | 43.84 | Artificial Analysis · GPQA Diamond |
| 496 | Devstral Small (May '25)Mistral | 43.43 | Artificial Analysis · GPQA Diamond |
| 497 | Nova LiteAmazon | 43.33 | Artificial Analysis · GPQA Diamond |
| 498 | Granite 4.1 8BIBM | 43.33 | Artificial Analysis · GPQA Diamond |
| 499 | Gemma 4 E2B (Reasoning)Google | 43.33 | Artificial Analysis · GPQA Diamond |
| 500 | Llama 3.2 Instruct 90B (Vision)Meta | 43.23 | Artificial Analysis · GPQA Diamond |
| 501 | GPT-5 nano (minimal)OpenAI | 42.83 | Artificial Analysis · GPQA Diamond |
| 502 | Gemma 3 27B InstructGoogle | 42.83 | Artificial Analysis · GPQA Diamond |
| 503 | Qwen3 VL 8B InstructAlibaba | 42.73 | Artificial Analysis · GPQA Diamond |
| 504 | Jamba 1.5 LargeAI21 Labs | 42.73 | Artificial Analysis · GPQA Diamond |
| 505 | GPT-4o miniOpenAI | 42.63 | Artificial Analysis · GPQA Diamond |
| 506 | Molmo2-8BAllen Institute for AI | 42.53 | Artificial Analysis · GPQA Diamond |
| 507 | Mistral SabaMistral | 42.42 | Artificial Analysis · GPQA Diamond |
| 508 | Exaone 4.0 1.2B (Non-reasoning)LG AI Research | 42.42 | Artificial Analysis · GPQA Diamond |
| 509 | DeepSeek-V2.5 (Dec '24)DeepSeek | 42.32 | Artificial Analysis · GPQA Diamond |
| 510 | Qwen2.5 Coder Instruct 32BAlibaba | 41.72 | Artificial Analysis · GPQA Diamond |
| 511 | Sarvam M (Reasoning)Sarvam | 41.62 | Artificial Analysis · GPQA Diamond |
| 512 | Granite 4.0 H SmallIBM | 41.62 | Artificial Analysis · GPQA Diamond |
| 513 | Devstral Small (Jul '25)Mistral | 41.41 | Artificial Analysis · GPQA Diamond |
| 514 | Kimi Linear 48B A3B InstructKimi | 41.21 | Artificial Analysis · GPQA Diamond |
| 515 | Qwen2.5 TurboAlibaba | 41.01 | Artificial Analysis · GPQA Diamond |
| 516 | Llama 3.1 Instruct 70BMeta | 40.91 | Artificial Analysis · GPQA Diamond |
| 517 | Claude 3.5 HaikuAnthropic | 40.81 | Artificial Analysis · GPQA Diamond |
| 518 | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)NVIDIA | 40.81 | Artificial Analysis · GPQA Diamond |
| 519 | Gemma 4 E2B (Non-reasoning)Google | 40.51 | Artificial Analysis · GPQA Diamond |
| 520 | DeepSeek R1 Distill Llama 70BDeepSeek | 40.2 | Artificial Analysis · GPQA Diamond |
| 521 | Hermes 3 - Llama-3.1 70BNous Research | 40.1 | Artificial Analysis · GPQA Diamond |
| 522 | Olmo 3 7B InstructAllen Institute for AI | 40 | Artificial Analysis · GPQA Diamond |
| 523 | Claude 3 SonnetAnthropic | 40 | Artificial Analysis · GPQA Diamond |
| 524 | NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA | 39.9 | Artificial Analysis · GPQA Diamond |
| 525 | Qwen3 4B (Non-reasoning)Alibaba | 39.8 | Artificial Analysis · GPQA Diamond |
| 526 | Jamba 1.7 LargeAI21 Labs | 38.99 | Artificial Analysis · GPQA Diamond |
| 527 | Jamba 1.6 LargeAI21 Labs | 38.69 | Artificial Analysis · GPQA Diamond |
| 528 | DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Nous Research | 38.18 | Artificial Analysis · GPQA Diamond |
| 529 | Mistral Small (Sep '24)Mistral | 38.08 | Artificial Analysis · GPQA Diamond |
| 530 | Llama 3 Instruct 70BMeta | 37.88 | Artificial Analysis · GPQA Diamond |
| 531 | Claude 3 HaikuAnthropic | 37.37 | Artificial Analysis · GPQA Diamond |
| 532 | Gemini 1.5 Pro (May '24)Google | 37.07 | Artificial Analysis · GPQA Diamond |
| 533 | Qwen2 Instruct 72BAlibaba | 37.07 | Artificial Analysis · GPQA Diamond |
| 534 | Qwen3 VL 4B InstructAlibaba | 37.07 | Artificial Analysis · GPQA Diamond |
| 535 | Gemini 1.5 Flash-8BGoogle | 35.86 | Artificial Analysis · GPQA Diamond |
| 536 | Nova MicroAmazon | 35.76 | Artificial Analysis · GPQA Diamond |
| 537 | Ministral 3 3BMistral | 35.76 | Artificial Analysis · GPQA Diamond |
| 538 | Qwen3 1.7B (Reasoning)Alibaba | 35.56 | Artificial Analysis · GPQA Diamond |
| 539 | Mistral Large (Feb '24)Mistral | 35.05 | Artificial Analysis · GPQA Diamond |
| 540 | Mistral MediumMistral | 34.95 | Artificial Analysis · GPQA Diamond |
| 541 | Gemma 3 12B InstructGoogle | 34.95 | Artificial Analysis · GPQA Diamond |
| 542 | GPT-4OpenAI | 34.95 | Artificial Analysis · GPQA Diamond |
| 543 | LFM2 8B A1BLiquid AI | 34.44 | Artificial Analysis · GPQA Diamond |
| 544 | Claude 2.0Anthropic | 34.44 | Artificial Analysis · GPQA Diamond |
| 545 | LFM2.5-1.2B-ThinkingLiquid AI | 33.94 | Artificial Analysis · GPQA Diamond |
| 546 | Qwen2.5 Coder Instruct 7BAlibaba | 33.94 | Artificial Analysis · GPQA Diamond |
| 547 | Granite 3.3 8B (Non-reasoning)IBM | 33.84 | Artificial Analysis · GPQA Diamond |
| 548 | Granite 4.0 MicroIBM | 33.64 | Artificial Analysis · GPQA Diamond |
| 549 | Jamba Reasoning 3BAI21 Labs | 33.33 | Artificial Analysis · GPQA Diamond |
| 550 | Mixtral 8x22B InstructMistral | 33.23 | Artificial Analysis · GPQA Diamond |
| 551 | DBRX InstructDatabricks | 33.13 | Artificial Analysis · GPQA Diamond |
| 552 | Phi-4 Mini InstructMicrosoft | 33.13 | Artificial Analysis · GPQA Diamond |
| 553 | Claude InstantAnthropic | 33.03 | Artificial Analysis · GPQA Diamond |
| 554 | OLMo 2 32BAllen Institute for AI | 32.83 | Artificial Analysis · GPQA Diamond |
| 555 | LFM 40BLiquid AI | 32.73 | Artificial Analysis · GPQA Diamond |
| 556 | Llama 2 Chat 70BMeta | 32.73 | Artificial Analysis · GPQA Diamond |
| 557 | LFM2.5-1.2B-InstructLiquid AI | 32.63 | Artificial Analysis · GPQA Diamond |
| 558 | Gemini 1.5 Flash (May '24)Google | 32.42 | Artificial Analysis · GPQA Diamond |
| 559 | Command-R+ (Apr '24)Cohere | 32.32 | Artificial Analysis · GPQA Diamond |
| 560 | Jamba 1.7 MiniAI21 Labs | 32.22 | Artificial Analysis · GPQA Diamond |
| 561 | Llama 2 Chat 13BMeta | 32.12 | Artificial Analysis · GPQA Diamond |
| 562 | Claude 2.1Anthropic | 31.92 | Artificial Analysis · GPQA Diamond |
| 563 | DeepSeek Coder V2 Lite InstructDeepSeek | 31.92 | Artificial Analysis · GPQA Diamond |
| 564 | Phi-3 Mini Instruct 3.8BMicrosoft | 31.92 | Artificial Analysis · GPQA Diamond |
| 565 | Phi-4 Multimodal InstructMicrosoft | 31.52 | Artificial Analysis · GPQA Diamond |
| 566 | Granite 4.1 3BIBM | 31.41 | Artificial Analysis · GPQA Diamond |
| 567 | LFM2 2.6BLiquid AI | 30.61 | Artificial Analysis · GPQA Diamond |
| 568 | MiniCPM-V 4.6 1.3BOpenBMB | 30.51 | Artificial Analysis · GPQA Diamond |
| 569 | Tiny Aya GlobalCohere | 30.51 | Artificial Analysis · GPQA Diamond |
| 570 | DeepSeek R1 Distill Llama 8BDeepSeek | 30.2 | Artificial Analysis · GPQA Diamond |
| 571 | Mistral Small (Feb '24)Mistral | 30.2 | Artificial Analysis · GPQA Diamond |
| 572 | Jamba 1.5 MiniAI21 Labs | 30.2 | Artificial Analysis · GPQA Diamond |
| 573 | Jamba 1.6 MiniAI21 Labs | 30 | Artificial Analysis · GPQA Diamond |
| 574 | GPT-3.5 TurboOpenAI | 29.7 | Artificial Analysis · GPQA Diamond |
| 575 | Llama 3 Instruct 8BMeta | 29.6 | Artificial Analysis · GPQA Diamond |
| 576 | Gemma 3n E4B InstructGoogle | 29.6 | Artificial Analysis · GPQA Diamond |
| 577 | Mixtral 8x7B InstructMistral | 29.19 | Artificial Analysis · GPQA Diamond |
| 578 | Gemma 3 4B InstructGoogle | 29.09 | Artificial Analysis · GPQA Diamond |
| 579 | LFM2.5-VL-1.6BLiquid AI | 28.89 | Artificial Analysis · GPQA Diamond |
| 580 | Qwen1.5 Chat 110BAlibaba | 28.89 | Artificial Analysis · GPQA Diamond |
| 581 | OLMo 2 7BAllen Institute for AI | 28.79 | Artificial Analysis · GPQA Diamond |
| 582 | Command-R (Mar '24)Cohere | 28.38 | Artificial Analysis · GPQA Diamond |
| 583 | Qwen3 1.7B (Non-reasoning)Alibaba | 28.28 | Artificial Analysis · GPQA Diamond |
| 584 | Granite 4.0 1BIBM | 28.08 | Artificial Analysis · GPQA Diamond |
| 585 | MiniCPM5-1B (Reasoning)OpenBMB | 27.78 | Artificial Analysis · GPQA Diamond |
| 586 | Gemma 3n E4B Instruct Preview (May '25)Google | 27.78 | Artificial Analysis · GPQA Diamond |
| 587 | Gemini 1.0 ProGoogle | 27.68 | Artificial Analysis · GPQA Diamond |
| 588 | Apertus 70B InstructSwiss AI Initiative | 27.17 | Artificial Analysis · GPQA Diamond |
| 589 | DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Nous Research | 26.97 | Artificial Analysis · GPQA Diamond |
| 590 | MiniCPM5-1B (Non-reasoning)OpenBMB | 26.87 | Artificial Analysis · GPQA Diamond |
| 591 | Granite 4.0 H 1BIBM | 26.26 | Artificial Analysis · GPQA Diamond |
| 592 | Granite 4.0 350MIBM | 26.06 | Artificial Analysis · GPQA Diamond |
| 593 | Llama 3.1 Instruct 8BMeta | 25.86 | Artificial Analysis · GPQA Diamond |
| 594 | Granite 4.0 H 350MIBM | 25.66 | Artificial Analysis · GPQA Diamond |
| 595 | Apertus 8B InstructSwiss AI Initiative | 25.56 | Artificial Analysis · GPQA Diamond |
| 596 | Llama 3.2 Instruct 3BMeta | 25.45 | Artificial Analysis · GPQA Diamond |
| 597 | Molmo 7B-DAllen Institute for AI | 24.04 | Artificial Analysis · GPQA Diamond |
| 598 | Qwen3 0.6B (Reasoning)Alibaba | 23.94 | Artificial Analysis · GPQA Diamond |
| 599 | Gemma 3 1B InstructGoogle | 23.74 | Artificial Analysis · GPQA Diamond |
| 600 | Qwen3.5 0.8B (Non-reasoning)Alibaba | 23.64 | Artificial Analysis · GPQA Diamond |
| 601 | Qwen3 0.6B (Non-reasoning)Alibaba | 23.13 | Artificial Analysis · GPQA Diamond |
| 602 | OpenChat 3.5 (1210)OpenChat | 23.03 | Artificial Analysis · GPQA Diamond |
| 603 | Gemma 3n E2B InstructGoogle | 22.93 | Artificial Analysis · GPQA Diamond |
| 604 | LFM2 1.2BLiquid AI | 22.83 | Artificial Analysis · GPQA Diamond |
| 605 | Llama 2 Chat 7BMeta | 22.73 | Artificial Analysis · GPQA Diamond |
| 606 | Gemma 3 270MGoogle | 22.42 | Artificial Analysis · GPQA Diamond |
| 607 | Llama 3.2 Instruct 11B (Vision)Meta | 22.12 | Artificial Analysis · GPQA Diamond |
| 608 | Llama 3.2 Instruct 1BMeta | 19.6 | Artificial Analysis · GPQA Diamond |
| 609 | Mistral 7B InstructMistral | 17.68 | Artificial Analysis · GPQA Diamond |
| 610 | Qwen3.5 0.8B (Reasoning)Alibaba | 11.11 | Artificial Analysis · GPQA Diamond |
| 611 | DeepSeek R1 Distill Qwen 1.5BDeepSeek | 9.8 | Artificial Analysis · GPQA Diamond |
일치하는 모델이 없습니다. 검색어를 줄여 보세요.
공개된 숫자만 높은 순으로 표시합니다. 순서가 신뢰구간을 고려한 확정 순위는 아닙니다. Artificial Analysis가 측정한 Diamond 결과만 사용합니다. 다른 회사의 자체 발표 점수는 섞지 않습니다. GPQA는 v4.2 종합 지수에서는 제외됐지만 개별 결과는 공개돼 있습니다.
평가 방법 확인 ↗읽는 법
전체 벤치마크전문가 수준 과학 문제를 푸는 reasoning 능력을 봅니다.
일상 대화 품질과는 다릅니다. 어려운 문제를 끝까지 버티는 모델을 볼 때 좋습니다.