Note: Overall leaderboard rankings may not reflect true model quality — individual benchmarks give a clearer picture. ARC-Challenge MMLU GPQA GSM8K Artificial Analysis Intelligence Index v4.1
← Back to leaderboard

Artificial Analysis Intelligence Index v4.0 (Maximum Reasoning)

255 models

Index, calculated on GDPval-AA, T2-Bench Telecom, Terminal Bench Hard, SciCode, AA-LCR, AA-Omniscience, IFBench, Humanity's Last Exam, GPQA Diamond, CritPt.

Top 5 Models Performance

Anthropic: Claude Fable 5 ######################################## 64.9
Anthropic: Claude Opus 4.8 ###################################### 61.4
OpenAI: GPT-5.5 ##################################### 60.2
Anthropic: Claude Opus 4.7 ################################### 57.3
Google: Gemini 3.1 Pro Preview ################################### 57.2
– 970 – 1.1T
– 2019 – 2026
Rank Model Score
🇺🇸 1 Anthropic: Claude Fable 5 64.9
🇺🇸 2 Anthropic: Claude Opus 4.8 61.4
🇺🇸 3 OpenAI: GPT-5.5 60.2
🇺🇸 4 Anthropic: Claude Opus 4.7 57.3
🇺🇸 5 Google: Gemini 3.1 Pro Preview 57.2
🇺🇸 6 OpenAI: GPT-5.4 56.8
🇨🇳 7 Qwen: Qwen3.7 Max通义千问 56.6
🇺🇸 8 Google: Gemini 3.5 Flash 55.3
🇨🇳 9 MiniMax: MiniMax M3稀宇科技 54.7
🇨🇳 10 Moonshot AI: Kimi K2.6月之暗面 53.9
🇨🇳 11 Xiaomi: MiMo-V2.5 Pro小米集团 53.8
🇺🇸 12 OpenAI: gpt-5.3-codex 53.6
🇨🇳 13 Qwen: qwen3.7-plus通义千问 53.3
🇺🇸 14 SpaceXAI: Grok 4.3 53.2
🇺🇸 15 Anthropic: claude-opus-4.6 52.9
🇺🇸 16 Meta: Muse Spark 52.2
🇨🇳 17 Qwen: qwen3.6-max通义千问 51.8
🇺🇸 18 Anthropic: Claude Sonnet 4.6 51.7
🇨🇳 19 DeepSeek: DeepSeek V4 Pro深度求索 51.5
🇨🇳 20 Z.ai: GLM 5.1北京智谱华章科技有限公司 51.4
🇺🇸 21 OpenAI: gpt-5.2 51.3
🇨🇳 22 Z.ai: GLM 5北京智谱华章科技有限公司 49.8
🇺🇸 23 Anthropic: claude-opus-4.5 49.7
🇨🇳 24 MiniMax: minimax-m2.7稀宇科技 49.6
🇨🇳 25 Xiaomi: mimo-v2-pro小米集团 49.2
🇨🇳 26 Xiaomi: MiMo-V2.5小米集团 49
🇺🇸 27 OpenAI: gpt-5.2-codex 49
🇺🇸 28 SpaceXAI: Grok 4.20 48.5
🇺🇸 29 Google: Gemini 3 Pro Preview 48.4
🇺🇸 30 NVIDIA Corporation: nvidia-nemotron-3-ultra 47.7
🇺🇸 31 OpenAI: gpt-5.1 47.7
🇨🇳 32 Moonshot AI: Kimi K2.5月之暗面 46.8
🇨🇳 33 DeepSeek: DeepSeek V4 Flash深度求索 46.5
🇨🇳 34 Qwen: qwen3.6-27b通义千问 45.8
🇨🇳 35 Qwen: qwen3.5-397b-a17b通义千问 45
🇺🇸 36 OpenAI: GPT-5 44.6
❓ 37 Kwaipilot: kat-coder-pro-v2 43.8
🇨🇳 38 Qwen: qwen3.6-35b-a3b通义千问 43.5
🇺🇸 39 Anthropic: claude-4.5-sonnet 43
🇨🇳 40 StepFun: step-3.7-flash上海阶跃星辰智能科技有限公司 42.6
🇨🇳 41 Z.ai: GLM 4.7北京智谱华章科技有限公司 42.1
🇨🇳 42 Qwen: Qwen3.5 27B通义千问 42.1
🇨🇳 43 MiniMax: MiniMax M2.5稀宇科技 41.9
🇨🇳 44 Tencent: Hy3 Preview腾讯控股有限公司 41.9
🇨🇳 45 DeepSeek: deepseek-v3.2深度求索 41.7
🇨🇳 46 Qwen: qwen3.5-122b-a10b通义千问 41.6
🇺🇸 47 SpaceXAI: Grok 4 41.5
🇺🇸 48 OpenAI: GPT-5 Mini 41.2
🇨🇳 49 Moonshot AI: Kimi K2 Thinking月之暗面 40.9
🇨🇳 50 MiniMax: MiniMax M2.1稀宇科技 39.4
🇺🇸 51 Google: Gemma 4 31B Instruct 39.2
🇫🇷 52 Mistral AI: mistral-medium-3.5-128b 39.2
🇺🇸 53 Anthropic: claude-sonnet-4 38.7
🇺🇸 54 SpaceXAI: Grok 4.1 Fast 38.6
🇨🇳 55 inclusionAI: ring-2.6-1t 38.5
🇺🇸 56 OpenAI: o3 38.4
🇨🇳 57 StepFun: step-3.5-flash上海阶跃星辰智能科技有限公司 37.8
🇨🇦 58 Cohere Labs: command-a-plus-05-2026-bf16 37.2
🇺🇸 59 Anthropic: claude-4.5-haiku 37.1
🇨🇳 60 Qwen: Qwen3.5 35B A3B通义千问 37.1
🇨🇳 61 MiniMax: minimax-m2稀宇科技 36.1
🇨🇳 62 China Mobile: jt-35b-flash中国移动通信集团公司 36.1
🇺🇸 63 NVIDIA Corporation: nvidia-nemotron-3-super-120b-a12b 36
❓ 64 Kwaipilot: kat-coder-pro-v1 36
🇺🇸 65 Amazon: nova-2.0-pro 35.7
🇺🇸 66 SpaceXAI: Grok 4 Fast 35.1
🇺🇸 67 Anthropic: claude-3.7-sonnet 34.7
🇺🇸 68 Google: gemini-2.5-pro 34.6
🇨🇳 69 inclusionAI: ling-2.6-1t 33.6
🇨🇳 70 ByteDance Seed: doubao-seed-code字节跳动 33.5
🇺🇸 71 OpenAI: gpt-oss-120b 33.3
🇺🇸 72 OpenAI: o4-mini 33.1
🇨🇳 73 DeepSeek: deepseek-v3.2-exp深度求索 32.9
🇺🇸 74 Inception: Mercury 2 32.8
🇨🇳 75 Qwen: Qwen3.5 4B通义千问 32.5
🇨🇳 76 Z.ai: glm-4.6北京智谱华章科技有限公司 32.5
🇨🇳 77 Qwen: Qwen3.5 9B通义千问 32.4
🇰🇷 78 LG AI Research: k-exaone-236b-a23b주식회사 엘지 32.1
🇺🇸 79 SpaceXAI: Grok 3 Mini 32.1
🇺🇸 80 Arcee AI: trinity-large-thinking 31.9
🇨🇳 81 Qwen: qwen3-max通义千问 31.4
🇺🇸 82 Google: Gemma 4 26B A4B Instruct 31.2
🇺🇸 83 OpenAI: o1 30.7
🇨🇳 84 Xiaomi: mimo-v2-flash小米集团 30.3
🇨🇳 85 Z.ai: glm-4.7-flash北京智谱华章科技有限公司 30.1
🇨🇳 86 DeepSeek: deepseek-v3.2-speciale深度求索 29.4
🇪🇸 87 Multiverse Computing: HyperNova 60B 29.3
🇺🇸 88 Google: Gemma 4 12B Instruct 29.2
🇨🇳 89 Baidu: ernie-5.0-thinking百度 29.1
🇺🇸 90 SpaceXAI: grok-code-fast-1 28.7
🇨🇳 91 Qwen: qwen3-coder-next通义千问 28.3
🇫🇷 92 Mistral AI: mistral-small-4-119b-2603 27.8
🇨🇦 93 Cohere Labs: north-mini-code-1.0 27.6
🇺🇸 94 OpenAI: GPT-5 Nano 26.8
🇨🇳 95 Qwen: qwen3-next-80b-a3b通义千问 26.7
🇺🇸 96 OpenAI: gpt-4.1 26.3
🇨🇳 97 Moonshot AI: Kimi K2 Instruct月之暗面 26.3
🇨🇳 98 inclusionAI: ling-2.6-flash 26.2
🇺🇸 99 OpenAI: o3-mini 25.9
❓ 100 upstage: solar-pro-3 25.9
🇺🇸 101 OpenAI: o1-pro 25.8
🇨🇳 102 China Mobile: jt-mini中国移动通信集团公司 25.4
🇺🇸 103 SpaceXAI: Grok 3 25.2
🇨🇳 104 ByteDance Seed: seed-oss-36b-instruct字节跳动 25.2
❓ 105 perplexity: sonar-reasoning-pro 24.6
🇺🇸 106 OpenAI: gpt-oss-20b 24.5
🇺🇸 107 OpenAI: gpt-5.4-nano 24.4
🇨🇳 108 MiniMax: minimax-m1-80k稀宇科技 24.4
🇺🇸 109 NVIDIA Corporation: nvidia-nemotron-3-nano-30b-a3b 24.3
🇨🇳 110 Meituan LongCat: longcat-flash-lite 23.9
❓ 111 naver-hyperclovax: hyperclovax-seed-think-32b 23.7
🇨🇳 112 Z.ai: glm-4.6v北京智谱华章科技有限公司 23.4
🇺🇸 113 OpenAI: gpt-5.4-mini 23.3
🇨🇳 114 Z.ai: glm-4.5-air北京智谱华章科技有限公司 23.2
🇨🇳 115 Qwen: Qwen3.5 2B通义千问 23
🇺🇸 116 OpenAI: gpt-4.1-mini 22.9
🇨🇳 117 inclusionAI: ring-1t 22.8
❓ 118 primeintellect: intellect-3 22.2
❓ 119 upstage: solar-open-100b 21.7
🇺🇸 120 SpaceXAI: Grok 3 Reasoning Beta 21.6
🇨🇳 121 MiniMax: minimax-m1-40k稀宇科技 20.9
🇨🇳 122 Qwen: qwq-32b通义千问 19.7
🇨🇳 123 inclusionAI: ling-1t 19
🇺🇸 124 Google: Gemma 4 E4B Instruct 18.8
🇨🇳 125 DeepSeek: DeepSeek R1深度求索 18.8
🇺🇸 126 Anthropic: claude-3.5-haiku 18.7
🇺🇸 127 OpenAI: gpt-4o 18.6
❓ 128 NousResearch: hermes-4-405b 18.6
🇺🇸 129 Meta Llama: Llama 4 Maverick Instruct 18.4
🇨🇳 130 OpenBMB: minicpm5-1b 18.2
🇮🇳 131 Sarvam AI: sarvam-105b 18.2
🇺🇸 132 Anthropic: claude-3-opus 18
❓ 133 perplexity: sonar-reasoning 17.9
🇺🇸 134 Meta Llama: Llama 3.1 405B Instruct 17.4
🇨🇳 135 DeepSeek: deepseek-r1-distill-qwen-32b深度求索 17.2
🇨🇳 136 Qwen: qwen3-vl-8b-thinking通义千问 16.7
🇰🇷 137 LG AI Research: exaone-4.0-32b주식회사 엘지 16.7
🇨🇳 138 DeepSeek: deepseek-v3深度求索 16.5
🇨🇳 139 Qwen: qwen3-32b通义千问 16.5
🇨🇳 140 DeepSeek: deepseek-r1-distill-qwen3-8b深度求索 16.4
🇨🇳 141 Qwen: qwen2.5-max通义千问 16.3
🇨🇳 142 Qwen: qwen3-14b通义千问 16.2
🇨🇳 143 DeepSeek: deepseek-r1-distill-llama-70b深度求索 16
🇫🇷 144 Mistral AI: ministral-3-14b-instruct-2512 16
❓ 145 NousResearch: hermes-4-70b 16
🇨🇳 146 Qwen: Qwen3.5 0.8B通义千问 15.9
🇨🇳 147 DeepSeek: deepseek-r1-distill-qwen-14b深度求索 15.8
🇦🇪 148 Technology Innovation Institute: falcon-h1r-7b 15.8
🇨🇳 149 Qwen: qwen2.5-72b-instruct通义千问 15.6
❓ 150 perplexity: sonar 15.5
🇨🇳 151 StepFun: step-3-vl-10b上海阶跃星辰智能科技有限公司 15.5
🇨🇳 152 Qwen: qwen3-30b-a3b通义千问 15.3
🇺🇸 153 Google: Gemma 4 E2B Instruct 15.2
❓ 154 perplexity: sonar-pro 15.2
🇨🇳 155 Z.ai: glm-4.5v北京智谱华章科技有限公司 15.1
🇨🇳 156 Baidu: ernie-4.5-300b-a47b百度 15
❓ 157 upstage: solar-pro-2 14.9
🇫🇷 158 Mistral AI: ministral-3-8b-instruct-2512 14.8
🇺🇸 159 NVIDIA Corporation: nvidia-nemotron-nano-9b-v2 14.8
🇺🇸 160 IBM Granite: granite-4.1-30b 14.7
🇺🇸 161 Meta Llama: Llama 3.3 70B Instruct 14.5
🇨🇳 162 Qwen: qwen3-vl-8b-instruct通义千问 14.3
🇨🇳 163 Qwen: Qwen3 4B通义千问 14.2
🇺🇸 164 LiquidAI: lfm-2.5-8b-a1b 14.2
🇺🇸 165 SpaceXAI: Grok 2 13.9
🇺🇸 166 Google: gemini-1.5-flash 13.8
🇺🇸 167 OpenAI: gpt-4-turbo 13.7
🇨🇳 168 Qwen: qwen3-vl-4b-thinking通义千问 13.7
🇨🇦 169 Cohere Labs: c4ai-command-a-03-2025 13.5
🇺🇸 170 Amazon: nova-pro-v1 13.5
🇨🇳 171 Qwen: Qwen3 8B通义千问 13.2
🇺🇸 172 OpenAI: gpt-4.1-nano 13
🇨🇳 173 Qwen: qwen2.5-coder-32b-instruct通义千问 12.9
🇺🇸 174 OpenAI: gpt-4 12.8
🇺🇸 175 Amazon: nova-lite-v1 12.7
🇨🇳 176 OpenBMB: minicpm-v-4.6 12.7
🇺🇸 177 IBM Granite: Granite 4.1 8B 12.4
🇺🇸 178 Anthropic: claude-3-haiku 12.3
🇨🇳 179 DeepSeek: deepseek-v2.5深度求索 12.3
🇮🇳 180 Sarvam AI: sarvam-30b 12.3
🇨🇳 181 DeepSeek: DeepSeek R1 Distill Llama 8B深度求索 12.1
❓ 182 perplexity: r1-1776 12
🇺🇸 183 Google: gemini-1.5-pro 12
🇺🇸 184 Reka AI: reka-flash 12
🇺🇸 185 Meta Llama: Llama 3.2 90B Instruct 11.9
❓ 186 upstage: solar-mini 11.9
🇺🇸 187 Meta Llama: Llama 3.1 8B Instruct 11.8
🇺🇸 188 SpaceXAI: Grok-1 11.7
🇨🇳 189 Qwen: qwen2-72b-instruct通义千问 11.7
🇫🇷 190 Mistral AI: ministral-3-3b-instruct-2512 11.2
❓ 191 NousResearch: deephermes-3-mistral-24b-preview 10.9
🇮🇱 192 AI21: jamba-1.7-large 10.9
🇮🇱 193 AI21: jamba-1.5-large 10.7
❓ 194 NousResearch: hermes-3-llama-3.1-70b 10.6
🇺🇸 195 Ai2: olmo-2-0325-32b-instruct 10.6
🇮🇱 196 AI21: jamba-1.6-large 10.6
🇺🇸 197 LiquidAI: lfm-2-24b-a2b 10.5
🇺🇸 198 Microsoft: phi-4 10.4
🇺🇸 199 Anthropic: claude-3-sonnet 10.3
🇺🇸 200 Google: Gemma 3 27B Instruct 10.3
🇺🇸 201 Amazon: nova-micro-v1 10.3
🇺🇸 202 Microsoft: phi-3-mini-4k-instruct 10.1
🇨🇳 203 Qwen: qwen2.5-coder-7b-instruct通义千问 10
🇺🇸 204 Microsoft: phi-4-multimodal-instruct 10
🇺🇸 205 Meta Llama: Llama 3.2 3B Instruct 9.7
🇺🇸 206 Meta Llama: llama-2-7b-chat 9.7
🇮🇱 207 AI21: jamba-reasoning-3b 9.6
🇨🇳 208 Qwen: qwen1.5-110b-chat通义千问 9.5
🇺🇸 209 Reka AI: reka-flash-3 9.5
🇺🇸 210 Ai2: olmo-3-7b-think 9.4
🇺🇸 211 Ai2: olmo-2-1124-7b-instruct 9.3
🇨🇳 212 DeepSeek: deepseek-r1-distill-qwen-1.5b深度求索 9.1
🇺🇸 213 OpenAI: GPT-3.5 Turbo 9
🇺🇸 214 Meta Llama: Llama 3 70B Instruct 8.9
🇨🇳 215 Qwen: qwen-chat-72b通义千问 8.8
🇺🇸 216 Google: Gemma 3 12B Instruct 8.8
🇺🇸 217 LiquidAI: lfm-40b 8.8
🇺🇸 218 Meta Llama: llama-3.2-11b-instruct 8.7
🇺🇸 219 Google: palm-2 8.6
🇺🇸 220 Snowflake: arctic-instruct 8.6
🇺🇸 221 IBM Granite: granite-4.1-3b 8.5
🇺🇸 222 Google: Gemini 1.0 Pro 8.5
🇨🇳 223 DeepSeek: deepseek-coder-v2-lite-instruct深度求索 8.5
🇺🇸 224 Microsoft: phi-4-mini-instruct 8.4
🇺🇸 225 Meta Llama: llama-2-70b-chat 8.4
🇺🇸 226 Meta Llama: llama-2-13b-chat 8.4
🇨🇳 227 DeepSeek: deepseek-llm-67b-chat深度求索 8.4
🇨🇦 228 Cohere Labs: c4ai-command-r-plus 8.3
🇰🇷 229 LG AI Research: exaone-4.0-1.2b주식회사 엘지 8.3
🇨🇳 230 OpenChat: openchat-3.5 8.3
🇺🇸 231 Databricks: DBRX Instruct 8.3
🇺🇸 232 LiquidAI: LFM2.5 1.2B Thinking 8.1
🇺🇸 233 Ai2: olmo-3-7b-instruct 8.1
🇮🇱 234 AI21: jamba-1.7-mini 8.1
🇨🇳 235 Qwen: qwen3-1.7b通义千问 8
🇺🇸 236 LiquidAI: LFM2.5 1.2B Instruct 8
🇮🇱 237 AI21: jamba-1.5-mini 8
🇺🇸 238 Google: Gemma 3 270M Instruct 7.7
🇨🇭 239 Swiss AI Initiative: apertus-70b-instruct-2509 7.7
🇨🇳 240 Qwen: qwen-chat-14b通义千问 7.4
🇨🇦 241 Cohere Labs: c4ai-command-r7b-12-2024 7.4
🇺🇸 242 Ai2: molmo2-8b 7.3
🇺🇸 243 LiquidAI: lfm-2-8b-a1b 7
🇮🇱 244 AI21: jamba-1.6-mini 7
🇨🇳 245 Qwen: Qwen3 0.6B通义千问 6.5
🇺🇸 246 Meta Llama: meta-llama-3-8b-instruct 6.4
🇺🇸 247 Google: Gemma 3n E4B Instruct 6.4
🇺🇸 248 Meta Llama: llama-3.2-1b-instruct 6.3
🇺🇸 249 Google: Gemma 3 4B Instruct 6.3
🇺🇸 250 IBM Granite: granite-4.0-350m 6.1
🇨🇭 251 Swiss AI Initiative: apertus-8b-instruct-2509 5.9
🇺🇸 252 Google: Gemma 3 1B Instruct 5.6
🇺🇸 253 IBM Granite: granite-4.0-h-350m 5.4
🇺🇸 254 Google: Gemma 3n E2B Instruct 4.8
🇨🇦 255 Cohere Labs: Tiny Aya Global 4.7