AI 大模型排行榜 (Artificial Analysis LLM Ranking)
信息查询
3.2k
次浏览
100%
有帮助
·
1 人反馈
值品工具箱提供的 AI 大模型排行榜聚合了来自 Artificial Analysis 的权威数据，实时追踪并排名超过 100 个主流大语言模型。
AI 大模型排行榜数据中心
| 排名 | 模型名称 | 综合指数 ▼ | 编程 | 价格 ($/1M) |
|---|---|---|---|---|
| 1 | Claude Opus 5 (Adaptive Reasoning, Max Effort) | 63.1 | 78 | $10 |
| 2 | Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) | 62.5 | 77 | $10 |
| 3 | Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) | 62.1 | 76.5 | $20 |
| 4 | Claude Opus 5 (Adaptive Reasoning, High Effort) | 61.5 | 76.5 | $10 |
| 5 | GPT-5.6 Sol (max) | 60.9 | 77.4 | $11.25 |
| 6 | Grok 4.6 (high) | 60.9 | 76.8 | $3 |
| 7 | Kimi K3 (max) | 59.7 | 76.2 | $6 |
| 8 | GPT-5.6 Sol (xhigh) | 59 | 78.3 | $11.25 |
| 9 | Claude Opus 5 (Adaptive Reasoning, Medium Effort) | 58.6 | 74.3 | $10 |
| 10 | Qwen3.8 Max | 58.1 | 71.8 | $3 |
| 11 | Qwen3.8 2.4T A95B | 57.7 | 71.9 | $3 |
| 12 | GPT-5.6 Sol (high) | 57.3 | 77.2 | $11.25 |
| 13 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) | 57.3 | 74.3 | $10 |
| 14 | Muse Spark 1.2 (xhigh) | 56.8 | 72.2 | $2 |
| 15 | GPT-5.6 Terra (max) | 56.6 | 76.7 | $4.5 |
| 16 | GPT-5.5 (xhigh) | 56.3 | 74.9 | $11.25 |
| 17 | Gemini 3.7 Flash (high) | 56 | 76.1 | $1.5 |
| 18 | Grok 4.5 (high) | 55.8 | 72.4 | $3 |
| 19 | GPT-5.6 Sol (medium) | 55.6 | 76.3 | $11.25 |
| 20 | Claude Sonnet 5 (Adaptive Reasoning, Max Effort) | 55.3 | 71.5 | $4 |
| 21 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort) | 55 | 73.6 | $10 |
| 22 | GPT-5.5 (high) | 54.7 | 71.6 | $11.25 |
| 23 | Gemini 3.7 Flash (medium) | 53.4 | 71.5 | $1.5 |
| 24 | DeepSeek V4 Pro 0813 (Reasoning, Max Effort) | 53.2 | 68.8 | $1.98 |
| 25 | Muse Spark 1.1 (xhigh) | 53.2 | 71.3 | $2 |
| 26 | GPT-5.4 (xhigh) | 53.1 | 71.1 | $5.625 |
| 27 | GPT-5.6 Terra (xhigh) | 52.8 | 70.6 | $4.5 |
| 28 | GLM-5.2 (max) | 52.6 | 68.8 | $2.15 |
| 29 | Claude Opus 5 (Adaptive Reasoning, Low Effort) | 52.5 | 66.9 | $10 |
| 30 | GPT-5.6 Luna (max) | 52.3 | 71.4 | $0.45 |
| 31 | Gemini 3.5 Flash (high) | 52 | 70.1 | $3.375 |
| 32 | DeepSeek V4 Flash 0731 (Reasoning, Max Effort) | 51.8 | 69.1 | $0.175 |
| 33 | Gemini 3.6 Flash (high) | 51.6 | 69.2 | $3 |
| 34 | GPT-5.5 (medium) | 51.4 | 71.5 | $11.25 |
| 35 | Gemini 3.7 Flash (low) | 50.9 | 71 | $1.5 |
| 36 | GPT-5.6 Sol (low) | 50.7 | 69.7 | $11.25 |
| 37 | GPT-5.6 Luna (xhigh) | 50.1 | 68.6 | $0.45 |
| 38 | GPT-5.6 Terra (high) | 50.1 | 67.1 | $4.5 |
| 39 | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) | 48.4 | 63 | $6 |
| 40 | Kimi K3 (low) | 48.3 | 72 | $6 |
| 41 | Gemini 3.1 Pro Preview | 47.7 | 68.8 | $4.5 |
| 42 | Motif 3 | 47.4 | 63.5 | $0 |
| 43 | GPT-5.6 Luna (high) | 47 | 63.3 | $0.45 |
| 44 | GPT-5.6 Terra (medium) | 46.8 | 64.7 | $4.5 |
| 45 | Gemini 3.5 Flash (medium) | 46.7 | - | $3.375 |
| 46 | Qwen3.7 Max | 46.7 | 66 | $3.75 |
| 47 | GPT-5.3 Codex (xhigh) | 45.5 | - | $4.813 |
| 48 | MiniMax-M3 | 45.4 | 58.6 | $0.525 |
| 49 | DeepSeek V4 Pro (Reasoning, Max Effort) | 45.3 | 59.4 | $0.544 |
| 50 | Motif 3 (Beta) | 45.3 | 62 | $0 |
| 51 | Kimi K2.6 | 45.1 | 61.8 | $1.712 |
| 52 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | 44.9 | - | $10 |
| 53 | GPT-5.5 (low) | 44.5 | 60.9 | $11.25 |
| 54 | Muse Spark | 44.3 | 58.6 | $0 |
| 55 | Claude Opus 4.7 (Non-reasoning, High Effort) | 43.9 | - | $10 |
| 56 | DeepSeek V4 Pro (Reasoning, High Effort) | 43.7 | 58.7 | $0.544 |
| 57 | GPT-5.2 (xhigh) | 43.3 | - | $4.813 |
| 58 | Kimi K2.7 Code | 43 | 60.8 | $1.712 |
| 59 | MiMo-V2.5-Pro | 42.9 | 60.2 | $0.544 |
| 60 | Claude Sonnet 5 (Non-reasoning, High Effort) | 42.6 | 66.4 | $4 |
| 61 | Inkling (xhigh) | 42.3 | 52.1 | $1.762 |
| 62 | Hy3 | 42.2 | 58.8 | $0.241 |
| 63 | Nex-N2-Pro | 42.1 | 59.1 | $1 |
| 64 | DeepSeek V4 Flash (Reasoning, Max Effort) | 42.1 | 56.2 | $0.168 |
| 65 | GPT-5.6 Sol (Non-reasoning) | 41.9 | 65.1 | $11.25 |
| 66 | Claude Opus 4.5 (Reasoning) | 41.9 | - | $10 |
| 67 | Solar Pro 4 | 41.6 | 52.7 | $0.525 |
| 68 | MiMo-V2-Pro | 41.4 | - | $0 |
| 69 | GPT-5.6 Terra (low) | 41.3 | 58.1 | $4.5 |
| 70 | Inkling Small | 41.2 | 52.9 | $0.525 |
| 71 | GPT-5.2 Codex (xhigh) | 41.2 | - | $4.813 |
| 72 | Qwen3.6 Max Preview | 41.1 | - | $2.925 |
| 73 | GLM-5.1 (Reasoning) | 41 | 55.8 | $2.135 |
| 74 | GPT-5.4 mini (xhigh) | 40.9 | 56.1 | $1.688 |
| 75 | Grok Build 0.1 0616 | 40.7 | 51.5 | $1.25 |
| 76 | Gemini 3 Pro Preview (high) | 40.6 | - | $4.5 |
| 77 | GLM-5 (Reasoning) | 40.6 | - | $1.55 |
| 78 | Qwen3.6 Plus | 40.5 | 54.5 | $1.125 |
| 79 | GPT-5.4 (low) | 40.2 | - | $5.625 |
| 80 | JT-4.1 Flash 236B A21B | 39.9 | 52.4 | $0 |
| 81 | Agnes 2.5 Pro Alpha | 39.7 | 58.8 | $0.563 |
| 82 | GPT-5.4 nano (xhigh) | 39.7 | 56.1 | $0.463 |
| 83 | Qwen3.7 Plus | 39.4 | 55.9 | $0.7 |
| 84 | GLM-5-Turbo | 39.1 | - | $0 |
| 85 | DeepSeek V4 Flash (Reasoning, High Effort) | 39 | 52 | $0.175 |
| 86 | GPT-5.6 Luna (medium) | 38.9 | 50.7 | $0.45 |
| 87 | GPT-5.2 (medium) | 38.9 | - | $4.813 |
| 88 | MiniMax-M2.7 | 38.9 | 52.6 | $0.525 |
| 89 | Claude Opus 4.6 (Non-reasoning, High Effort) | 38.8 | - | $10 |
| 90 | Gemini 3 Flash Preview (Reasoning) | 38.7 | - | $1.125 |
| 91 | Nemotron 3 Ultra 550B A55B (Reasoning) | 38.3 | 49.3 | $1.137 |
| 92 | MiMo-V2.5 | 38 | 56.8 | $0.175 |
| 93 | Grok 4.20 0309 v2 (Reasoning) | 38 | - | $1.563 |
| 94 | Grok 4.3 (high) | 37.9 | 42.2 | $1.563 |
| 95 | Ling 3.0 Flash | 37.8 | 50.6 | $0.111 |
| 96 | Qwen3.6 27B (Reasoning) | 37.7 | 53.7 | $1.35 |
| 97 | GPT-5.1 (high) | 37.5 | 49.4 | $3.438 |
| 98 | Gemini 3.5 Flash-Lite | 37.4 | 49.3 | $0.85 |
| 99 | Solar Open2 250B | 37.4 | 44.7 | $0 |
| 100 | Claude 4.5 Sonnet (Reasoning) | 37.4 | 52.1 | $6 |
| 101 | Grok 4.20 0309 (Reasoning) | 37.4 | - | $3 |
| 102 | MiMo-V2-Omni-0327 | 37.3 | - | $0 |
| 103 | GPT-5 Codex (high) | 37 | - | $3.438 |
| 104 | Grok 4.3 (medium) | 36.9 | - | $1.563 |
| 105 | Claude Sonnet 4.6 (Non-reasoning, High Effort) | 36.8 | - | $6 |
| 106 | Grok 4.3 (low) | 36.3 | - | $1.563 |
| 107 | GLM-5.1 (Non-reasoning) | 36.3 | - | $2.135 |
| 108 | Kimi K2.5 (Reasoning) | 36 | 46.8 | $1.2 |
| 109 | MiMo-V2-Omni | 35.9 | - | $0 |
| 110 | Gemini 3.5 Flash (minimal) | 35.8 | - | $3.375 |
| 111 | GPT-5.5 (Non-reasoning) | 35.8 | 56.5 | $11.25 |
| 112 | GPT-5.1 Codex (high) | 35.6 | - | $3.438 |
| 113 | Claude Opus 4.5 (Non-reasoning) | 35.6 | - | $10 |
| 114 | Kimi K2.6 (Non-reasoning) | 35.4 | - | $1.712 |
| 115 | GPT-5 (high) | 35.3 | 37.8 | $3.438 |
| 116 | GLM 5V Turbo (Reasoning) | 35.3 | - | $0 |
| 117 | Muse Glimmer (high) | 35.1 | 49 | $0.581 |
| 118 | Claude Sonnet 4.6 (Non-reasoning, Low Effort) | 35.1 | - | $6 |
| 119 | A.X-K2 | 35 | 38.8 | $0 |
| 120 | GLM-5.2 (Non-reasoning) | 34.8 | 46.5 | $2.15 |
| 121 | GPT-5.6 Terra (Non-reasoning) | 34.6 | 52.3 | $4.5 |
| 122 | GPT-5 (medium) | 34.6 | - | $3.438 |
| 123 | Qwen3.5 27B (Reasoning) | 34.6 | - | $0.825 |
| 124 | KAT Coder Pro V2 | 34.5 | 59.5 | $0.525 |
| 125 | Claude 4.1 Opus (Reasoning) | 34.5 | - | $30 |
| 126 | MiniMax-M2.5 | 34.5 | - | $0.525 |
| 127 | GLM-4.7 (Reasoning) | 34.5 | 45.3 | $1 |
| 128 | Hy3-preview (Reasoning) | 34.4 | - | $0.1 |
| 129 | LongCat 2.0 | 34.3 | 45.3 | $1.3 |
| 130 | Qwen3.5 397B A17B (Reasoning) | 34.3 | 48.2 | $1.35 |
| 131 | GPT-5.5 Instant (May 2026) | 34.3 | - | $11.25 |
| 132 | Grok 4 | 34.1 | - | $6 |
| 133 | MiMo-V2-Flash (Feb 2026) | 34 | - | $0 |
| 134 | GPT-5.6 Luna (low) | 33.9 | 44.2 | $0.45 |
| 135 | Gemini 3 Pro Preview (low) | 33.9 | - | $4.5 |
| 136 | Kimi K2 Thinking | 33.5 | - | $1.075 |
| 137 | o3-pro | 33.3 | - | $35 |
| 138 | GLM-5 (Non-reasoning) | 33.2 | - | $1.55 |
| 139 | Qwen3.5 122B A10B (Reasoning) | 32.8 | 45.7 | $1.1 |
| 140 | DeepSeek V3.2 (Reasoning) | 32.8 | 44.2 | $0.315 |
| 141 | Qwen3.5 397B A17B (Non-reasoning) | 32.7 | - | $1.35 |
| 142 | Qwen3 Max Thinking | 32.5 | - | $0 |
| 143 | Qwen3.6 35B A3B (Reasoning) | 32.1 | 41.9 | $0.844 |
| 144 | MiniMax-M2.1 | 32.1 | - | $0.525 |
| 145 | DeepSeek V4 Pro (Non-reasoning) | 31.9 | - | $0.544 |
| 146 | GPT-5 (low) | 31.9 | - | $3.438 |
| 147 | MiMo-V2-Flash (Reasoning) | 31.9 | - | $0.15 |
| 148 | Claude 4 Opus (Reasoning) | 31.7 | - | $30 |
| 149 | G9v3-39A5B | 31.6 | 31.7 | $0 |
| 150 | GPT-5 mini (medium) | 31.6 | - | $0.688 |
| 151 | Qwen3.6 27B (Non-reasoning) | 31.3 | 46.6 | $1.35 |
| 152 | Qwen3.5 Omni Plus | 31.3 | - | $1.5 |
| 153 | Ring-2.6-1T | 31.3 | 42.8 | $0.85 |
| 154 | GPT-5.1 Codex mini (high) | 31.3 | - | $0.688 |
| 155 | Grok 4.1 Fast (Reasoning) | 31.3 | - | $0 |
| 156 | o3 | 31.1 | - | $3.5 |
| 157 | DeepSeek V3.1 Terminus (Reasoning) | 31.1 | 43.5 | $1.914 |
| 158 | K-EXAONE 2.0 0803 | 31 | 40.6 | $0 |
| 159 | Step 3.7 Flash | 30.9 | 39.6 | $0.438 |
| 160 | GPT-5.4 nano (medium) | 30.8 | - | $0.463 |
| 161 | GPT-5.4 mini (medium) | 30.5 | - | $1.688 |
| 162 | Mistral Medium 3.5 | 30.4 | 46.9 | $3 |
| 163 | Kimi K2.5 (Non-reasoning) | 30.1 | - | $1.2 |
| 164 | Qwen3.5 27B (Non-reasoning) | 30 | - | $0.825 |
| 165 | Claude 4.5 Haiku (Reasoning) | 29.9 | 43.9 | $2 |
| 166 | Claude 4.5 Sonnet (Non-reasoning) | 29.9 | - | $6 |
| 167 | Qwen3.5 35B A3B (Reasoning) | 29.9 | - | $0.688 |
| 168 | Claude 4 Sonnet (Reasoning) | 29.8 | 37.6 | $6 |
| 169 | Gemma 4 31B (Reasoning) | 29.7 | 43.4 | $0 |
| 170 | DeepSeek V4 Flash (Non-reasoning) | 29.3 | - | $0.175 |
| 171 | GLM-4.6 (Reasoning) | 29.3 | 45.8 | $0.963 |
| 172 | GPT-5.5 Instant (June 2026) | 29.2 | 39.4 | $11.25 |
| 173 | JT-35B-Flash | 29 | - | $0 |
| 174 | MiniMax-M2 | 28.9 | - | $0.525 |
| 175 | KAT-Coder-Pro V1 | 28.9 | - | $0 |
| 176 | Claude 4.1 Opus (Non-reasoning) | 28.8 | - | $30 |
| 177 | MiMo-V2.5-Pro (Non-reasoning) | 28.4 | - | $0.544 |
| 178 | GPT-5.4 (Non-reasoning) | 28.3 | - | $5.625 |
| 179 | Qwen3.5 122B A10B (Non-reasoning) | 28.2 | 43.3 | $1.1 |
| 180 | Gemini 3 Flash Preview (Non-reasoning) | 27.9 | - | $1.125 |
| 181 | Grok 4 Fast (Reasoning) | 27.9 | - | $0.275 |
| 182 | Claude 3.7 Sonnet (Reasoning) | 27.6 | 36.4 | $0 |
| 183 | GLM-4.7 (Non-reasoning) | 27.1 | - | $1 |
| 184 | GPT-5.6 Luna (Non-reasoning) | 26.8 | 39.3 | $0.45 |
| 185 | Hy3-preview (Non-reasoning) | 26.6 | - | $0.1 |
| 186 | Ling-2.6-1T | 26.6 | - | $0.85 |
| 187 | Doubao Seed Code | 26.5 | - | $0 |
| 188 | GPT-5.2 (Non-reasoning) | 26.5 | - | $4.813 |
| 189 | Step 3.5 Flash 2603 | 26.5 | - | $0.15 |
| 190 | Gemma 4 26B A4B (Reasoning) | 26.1 | 39.3 | $0.198 |
| 191 | o4-mini (high) | 26.1 | - | $1.925 |
| 192 | Claude 4 Opus (Non-reasoning) | 26 | - | $30 |
| 193 | Claude 4 Sonnet (Non-reasoning) | 26 | - | $6 |
| 194 | Step 3.5 Flash | 26 | - | $0.15 |
| 195 | Gemini 2.5 Pro | 25.9 | 33.3 | $3.438 |
| 196 | DeepSeek V3.2 Exp (Reasoning) | 25.9 | - | $0.315 |
| 197 | GPT-5 mini (high) | 25.8 | 15.6 | $0.688 |
| 198 | Nemotron 3 Super 120B A12B (Reasoning) | 25.7 | 37.7 | $0.35 |
| 199 | Gemini 3.1 Flash-Lite | 25.6 | 34.7 | $0.563 |
| 200 | Qwen3 Max Thinking (Preview) | 25.5 | - | $2.4 |
| 201 | MiMo-V2-Flash (Non-reasoning) | 25.1 | 49.8 | $0 |
| 202 | DeepSeek V3.2 (Non-reasoning) | 25.1 | - | $0.315 |
| 203 | Grok 4.3 (Non-reasoning) | 25 | 35.2 | $1.563 |
| 204 | Qwen3.6 35B A3B (Non-reasoning) | 24.6 | 28.1 | $0.844 |
| 205 | Ling 3.0 Tiny | 24.5 | 26.5 | $0 |
| 206 | Qwen3 Max | 24.5 | - | $2.4 |
| 207 | Qwen3.5 35B A3B (Non-reasoning) | 24.3 | 37 | $0.688 |
| 208 | Gemini 2.5 Flash Preview (Sep '25) (Reasoning) | 24.2 | - | $0 |
| 209 | gpt-oss-120b (high) | 24.1 | 30.4 | $0.262 |
| 210 | Claude 4.5 Haiku (Non-reasoning) | 24.1 | - | $2 |
| 211 | Kimi K2 0905 | 24 | - | $1.075 |
| 212 | o1 | 23.9 | 39.7 | $26.25 |
| 213 | Claude 3.7 Sonnet (Non-reasoning) | 23.9 | - | $6 |
| 214 | Nemotron 3.5 Lightning | 23.6 | 26.8 | $0.088 |
| 215 | Gemini 2.5 Pro Preview (Mar' 25) | 23.4 | 46.7 | $0 |
| 216 | GLM-4.6 (Non-reasoning) | 23.4 | - | $0.981 |
| 217 | GLM-4.7-Flash (Reasoning) | 23.3 | - | $0.153 |
| 218 | Grok 3 mini Reasoning (high) | 22.9 | - | $0.35 |
| 219 | Grok 4.20 0309 (Non-reasoning) | 22.9 | - | $3 |
| 220 | Command A+ | 22.8 | 27.8 | $0 |
| 221 | Gemini 2.5 Pro Preview (May' 25) | 22.7 | - | $3.438 |
| 222 | DeepSeek V3.2 Speciale | 22.6 | - | $0 |
| 223 | K-EXAONE (Reasoning) | 22.5 | 32.1 | $0 |
| 224 | Gemma 4 31B (Non-reasoning) | 22.3 | 33.2 | $0.209 |
| 225 | ERNIE 5.0 Thinking Preview | 22.3 | - | $0 |
| 226 | Gemma 4 12B (Reasoning) | 22.2 | 31 | $0.15 |
| 227 | Grok 4.20 0309 v2 (Non-reasoning) | 22.2 | - | $1.563 |
| 228 | Nova 2.0 Pro Preview (medium) | 22.1 | 34 | $3.438 |
| 229 | Grok Code Fast 1 | 22 | - | $0 |
| 230 | Mercury 2 | 21.9 | 31.1 | $0.375 |
| 231 | Qwen3.5 9B (Reasoning) | 21.8 | 28.7 | $0.151 |
| 232 | DeepSeek V3.1 Terminus (Non-reasoning) | 21.7 | - | $0.453 |
| 233 | DeepSeek V3.2 Exp (Non-reasoning) | 21.7 | - | $0.315 |
| 234 | Apriel-v1.5-15B-Thinker | 21.6 | - | $0 |
| 235 | DeepSeek V3.1 (Non-reasoning) | 21.4 | - | $0.84 |
| 236 | Nova 2.0 Omni (medium) | 21.3 | - | $0.85 |
| 237 | Qwen3 Coder Next | 21.3 | 36.2 | $0.563 |
| 238 | DeepSeek V3.1 (Reasoning) | 21 | - | $0.865 |
| 239 | Qwen3 VL 235B A22B (Reasoning) | 20.9 | - | $2.625 |
| 240 | Nova 2.0 Lite (high) | 20.8 | 23 | $0.85 |
| 241 | Apriel-v1.6-15B-Thinker | 20.8 | - | $0 |
| 242 | GPT-5.1 (Non-reasoning) | 20.7 | - | $3.438 |
| 243 | Qwen3.5 9B (Non-reasoning) | 20.6 | 23.5 | $0.19 |
| 244 | EXAONE 4.5 33B | 20.5 | 23.6 | $0 |
| 245 | Gemma 4 26B A4B (Non-reasoning) | 20.4 | - | $0.198 |
| 246 | Qwen3.5 4B (Reasoning) | 20.4 | 22.6 | $0.06 |
| 247 | DeepSeek R1 0528 (May '25) | 20.4 | - | $2.063 |
| 248 | Gemini 2.5 Flash (Reasoning) | 20.3 | - | $0.85 |
| 249 | North Mini Code | 20.2 | 36.5 | $0 |
| 250 | GPT-5 nano (high) | 20.1 | - | $0.138 |
| 251 | Qwen3 235B A22B 2507 (Reasoning) | 19.9 | 22.1 | $2.625 |
| 252 | Nova 2.0 Pro Preview (low) | 19.8 | 25.9 | $3.438 |
| 253 | Mistral Small 4 (Reasoning) | 19.7 | 26.6 | $0.262 |
| 254 | Kimi K2 | 19.7 | - | $1.002 |
| 255 | GLM-4.5 (Reasoning) | 19.7 | - | $0 |
| 256 | GPT-4.1 | 19.6 | - | $3.5 |
| 257 | Qwen3 Max (Preview) | 19.4 | - | $2.4 |
| 258 | Devstral 2 | 19.2 | 31.3 | $0 |
| 259 | Nova 2.0 Lite (medium) | 19.2 | - | $0.85 |
| 260 | Qwen3.5 Omni Flash | 19.2 | - | $0.275 |
| 261 | o3-mini | 19.2 | - | $1.925 |
| 262 | GPT-5 nano (medium) | 19.2 | - | $0.138 |
| 263 | o1-pro | 19.1 | - | $262.5 |
| 264 | Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning) | 19.1 | - | $0 |
| 265 | JT-MINI | 18.8 | - | $0 |
| 266 | DeepSeek R1 (Jan '25) | 18.6 | 24.6 | $2.431 |
| 267 | Grok 3 | 18.6 | - | $8 |
| 268 | Seed-OSS-36B-Instruct | 18.5 | - | $0.3 |
| 269 | Trinity Large Thinking | 18.4 | 25.8 | $0.395 |
| 270 | Qwen3 235B A22B 2507 Instruct | 18.4 | - | $1.225 |
| 271 | HyperNova 60B 2605 | 18.3 | 23.2 | $0 |
| 272 | Qwen3 Coder 480B A35B Instruct | 18.2 | - | $3 |
| 273 | Qwen3 VL 32B (Reasoning) | 18.1 | - | $2.625 |
| 274 | Magistral Medium 1.2 | 18 | 21.3 | $2.75 |
| 275 | Nova 2.0 Lite (low) | 18 | - | $0.85 |
| 276 | Sonar Reasoning Pro | 18 | - | $0 |
| 277 | MiniMax M1 80k | 17.9 | - | $0.963 |
| 278 | Nemotron Cascade 2 30B A3B | 17.8 | 25.3 | $0 |
| 279 | GPT-5.4 nano (Non-Reasoning) | 17.8 | - | $0.463 |
| 280 | Devstral Small 2 | 17.7 | 29.3 | $0 |
| 281 | Gemini 2.5 Flash Preview (Reasoning) | 17.7 | - | $0 |
| 282 | K2 Think V2 | 17.4 | 21 | $0 |
| 283 | LongCat Flash Lite | 17.4 | - | $0 |
| 284 | GPT-5 (minimal) | 17.3 | - | $3.438 |
| 285 | HyperCLOVA X SEED Think (32B) | 17.2 | - | $0 |
| 286 | o1-preview | 17.2 | 34 | $28.875 |
| 287 | Grok 4.1 Fast (Non-reasoning) | 17 | - | $0 |
| 288 | K-EXAONE (Non-reasoning) | 16.9 | - | $0 |
| 289 | Qwen3 Next 80B A3B (Reasoning) | 16.9 | 17.4 | $1.875 |
| 290 | GLM-4.6V (Reasoning) | 16.9 | - | $0.45 |
| 291 | GPT-5.4 mini (Non-Reasoning) | 16.8 | - | $1.688 |
| 292 | Nova 2.0 Omni (low) | 16.7 | - | $0.85 |
| 293 | GLM-4.5-Air | 16.7 | - | $0.372 |
| 294 | Mi:dm K 2.5 Pro | 16.6 | - | $0 |
| 295 | Grok 4 Fast (Non-reasoning) | 16.6 | - | $0.275 |
| 296 | Ring-1T | 16.3 | - | $0 |
| 297 | G9v3-3B | 16.2 | 9.9 | $0 |
| 298 | Qwen3.5 4B (Non-reasoning) | 16.1 | 20.3 | $0.06 |
| 299 | Mistral Large 3 | 15.9 | 20.1 | $0.75 |
| 300 | INTELLECT-3 | 15.7 | - | $0 |
| 301 | o3-mini (high) | 15.7 | 16.3 | $1.925 |
| 302 | GLM-4.7-Flash (Non-reasoning) | 15.6 | - | $0.153 |
| 303 | GPT-5 (ChatGPT) | 15.4 | - | $0 |
| 304 | gpt-oss-20b (high) | 15.2 | 20.7 | $0.092 |
| 305 | Solar Open 100B (Reasoning) | 15.2 | - | $0 |
| 306 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) | 15.2 | - | $0.175 |
| 307 | DeepSeek V3 0324 | 15.2 | 21.2 | $0.483 |
| 308 | Grok 3 Reasoning Beta | 15.2 | - | $0 |
| 309 | Nemotron 3 Nano Omni 30B A3B Reasoning | 15 | 13.8 | $0.131 |
| 310 | gpt-oss-120b (low) | 14.9 | 21.2 | $0.261 |
| 311 | Mistral Small 3.1 | 14.9 | 26.3 | $0.15 |
| 312 | GPT-4.1 mini | 14.8 | 20.2 | $0.7 |
| 313 | Mistral Medium 3.1 | 14.7 | 20.5 | $0.8 |
| 314 | Qwen3 30B A3B 2507 (Reasoning) | 14.6 | 12.1 | $0.75 |
| 315 | Llama 4 Maverick | 14.5 | 16.3 | $0.415 |
| 316 | Solar Pro 3 | 14.5 | 16.2 | $0.262 |
| 317 | NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) | 14.5 | 14.4 | $0.088 |
| 318 | MiniMax M1 40k | 14.5 | - | $0 |
| 319 | gpt-oss-20b (low) | 14.4 | - | $0.103 |
| 320 | Nova 2.0 Pro Preview (Non-reasoning) | 14.4 | 20.9 | $3.438 |
| 321 | Qwen3 VL 235B A22B Instruct | 14.4 | - | $1.225 |
| 322 | GPT-5 mini (minimal) | 14.3 | - | $0.688 |
| 323 | K2-V2 (high) | 14.2 | - | $0 |
| 324 | Gemini 2.5 Flash (Non-reasoning) | 14.2 | - | $0.85 |
| 325 | DeepSeek V3 (Dec '24) | 14.2 | 23 | $0.493 |
| 326 | Ling 2.6 Flash | 14.2 | 25.3 | $0.15 |
| 327 | o1-mini | 14 | - | $0 |
| 328 | Qwen3 Next 80B A3B Instruct | 13.8 | - | $0.875 |
| 329 | GPT-4.5 (Preview) | 13.6 | - | $0 |
| 330 | Tri-21B-think Preview | 13.6 | - | $0 |
| 331 | Qwen3 Coder 30B A3B Instruct | 13.6 | - | $0.9 |
| 332 | DiffusionGemma 26B A4B | 13.5 | 19.7 | $0 |
| 333 | Qwen3 235B A22B (Reasoning) | 13.5 | - | $2.625 |
| 334 | QwQ 32B | 13.4 | - | $0.745 |
| 335 | Qwen3 VL 30B A3B (Reasoning) | 13.4 | - | $0.75 |
| 336 | Gemini 2.0 Flash Thinking Experimental (Jan '25) | 13.3 | 24.1 | $0 |
| 337 | Gemma 4 12B (Non-reasoning) | 13.2 | - | $0.15 |
| 338 | Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) | 13.1 | - | $0.175 |
| 339 | Motif-2-12.7B-Reasoning | 12.8 | - | $0 |
| 340 | Nova Premier | 12.7 | - | $5 |
| 341 | Ling-1T | 12.7 | - | $0 |
| 342 | Mistral Medium 3 | 12.5 | - | $0.8 |
| 343 | Magistral Medium 1 | 12.5 | - | $0 |
| 344 | Solar Pro 2 (Preview) (Reasoning) | 12.5 | - | $0 |
| 345 | Llama Nemotron Super 49B v1.5 (Reasoning) | 12.4 | - | $0.4 |
| 346 | K2-V2 (medium) | 12.4 | - | $0 |
| 347 | Celeris-1 | 12.4 | 14.4 | $0.325 |
| 348 | Devstral Medium | 12.4 | - | $0 |
| 349 | Mistral Small 4 (Non-reasoning) | 12.3 | - | $0.262 |
| 350 | Tri-21B-Think | 12.3 | - | $0 |
| 351 | GPT-4o (March 2025, chatgpt-4o-latest) | 12.3 | - | $0 |
| 352 | Gemma 4 E4B (Reasoning) | 12.2 | 9.4 | $0.04 |
| 353 | Gemini 2.0 Flash (Feb '25) | 12.2 | - | $0 |
| 354 | Claude 3.5 Haiku | 12.2 | 15.9 | $0 |
| 355 | Llama 3.3 Nemotron Super 49B v1 (Reasoning) | 12.2 | - | $0 |
| 356 | Sarvam 105B (high) | 11.9 | - | $0.074 |
| 357 | MiniCPM5-1B (Reasoning) | 11.9 | - | $0 |
| 358 | Qwen3 4B 2507 (Reasoning) | 11.9 | - | $0 |
| 359 | Nova 2.0 Lite (Non-reasoning) | 11.8 | - | $0.85 |
| 360 | Gemini 2.0 Pro Experimental (Feb '25) | 11.8 | 25.5 | $0 |
| 361 | Claude 3 Opus | 11.8 | 19.5 | $30 |
| 362 | Devstral Small (May '25) | 11.8 | - | $0 |
| 363 | MiniCPM5-1B (Non-reasoning) | 11.7 | - | $0 |
| 364 | Gemini 2.5 Flash Preview (Non-reasoning) | 11.6 | - | $0 |
| 365 | Sonar Reasoning | 11.6 | - | $0 |
| 366 | Magistral Small 1.2 | 11.5 | 14.7 | $0.75 |
| 367 | Gemini 2.5 Flash-Lite (Reasoning) | 11.4 | - | $0.175 |
| 368 | Qwen3 32B (Reasoning) | 11.4 | 15.3 | $2.625 |
| 369 | Ministral 3 14B | 11.2 | 14.4 | $0.2 |
| 370 | GPT-4o (Nov '24) | 11.1 | - | $4.375 |
| 371 | Nanbeige4.1-3B | 11 | 9.6 | $0 |
| 372 | DeepSeek R1 Distill Qwen 32B | 11 | - | $0 |
| 373 | Qwen3 VL 32B Instruct | 11 | - | $1.225 |
| 374 | GLM-4.6V (Non-reasoning) | 10.9 | - | $0.45 |
| 375 | Qwen3 235B A22B (Non-reasoning) | 10.8 | - | $1.225 |
| 376 | Mistral Small 3.2 | 10.7 | 12.5 | $0.15 |
| 377 | Gemini 2.0 Flash (experimental) | 10.6 | - | $0 |
| 378 | Magistral Small 1 | 10.6 | - | $0 |
| 379 | EXAONE 4.0 32B (Reasoning) | 10.5 | - | $0 |
| 380 | Qwen3 VL 8B (Reasoning) | 10.5 | - | $0.66 |
| 381 | Nova 2.0 Omni (Non-reasoning) | 10.4 | - | $0.85 |
| 382 | Qwen3 14B (Reasoning) | 10.4 | 13.8 | $1.313 |
| 383 | Llama 4 Scout | 10.3 | 8.2 | $0.3 |
| 384 | DeepSeek R1 0528 Qwen3 8B | 10.3 | - | $0 |
| 385 | Qwen2.5 Max | 10.1 | - | $0 |
| 386 | Hermes 4 - Llama-3.1 70B (Reasoning) | 9.9 | - | $0.198 |
| 387 | Gemini 1.5 Pro (Sep '24) | 9.9 | 23.6 | $0 |
| 388 | Solar Pro 2 (Preview) (Non-reasoning) | 9.9 | - | $0 |
| 389 | Qwen3 VL 30B A3B Instruct | 9.9 | - | $0.35 |
| 390 | Claude 3.5 Sonnet (Oct '24) | 9.8 | 30.2 | $6 |
| 391 | DeepSeek R1 Distill Llama 70B | 9.8 | - | $0.8 |
| 392 | Falcon-H1R-7B | 9.7 | - | $0 |
| 393 | DeepSeek R1 Distill Qwen 14B | 9.7 | - | $0 |
| 394 | GPT-4.1 nano | 9.6 | 11.1 | $0.175 |
| 395 | Ling-flash-2.0 | 9.6 | - | $0.247 |
| 396 | Gemma 4 E2B (Reasoning) | 9.5 | 7.2 | $0 |
| 397 | Qwen3 Omni 30B A3B (Reasoning) | 9.5 | - | $0.43 |
| 398 | GPT-4o (Aug '24) | 9.4 | - | $4.375 |
| 399 | Sonar | 9.4 | - | $0 |
| 400 | Qwen2.5 Instruct 72B | 9.4 | - | $0.48 |
| 401 | Llama 3.3 Instruct 70B | 9.3 | 11.9 | $0.671 |
| 402 | Step3 VL 10B | 9.3 | - | $0 |
| 403 | Qwen3 30B A3B (Reasoning) | 9.2 | - | $0.75 |
| 404 | Devstral Small (Jul '25) | 9.1 | - | $0 |
| 405 | Sonar Pro | 9.1 | - | $0 |
| 406 | QwQ 32B-Preview | 9.1 | - | $0 |
| 407 | Ministral 3 8B | 9 | 9.7 | $0.15 |
| 408 | Mistral Large 2 (Nov '24) | 9 | - | $0 |
| 409 | GLM-4.5V (Reasoning) | 9 | - | $0.9 |
| 410 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning) | 8.9 | - | $0.9 |
| 411 | ERNIE 4.5 300B A47B | 8.9 | - | $0.485 |
| 412 | Qwen3 30B A3B 2507 Instruct | 8.9 | - | $0.35 |
| 413 | NVIDIA Nemotron Nano 12B v2 VL (Reasoning) | 8.8 | - | $0.3 |
| 414 | Hermes 4 - Llama-3.1 405B (Reasoning) | 8.8 | - | $1.5 |
| 415 | Solar Pro 2 (Reasoning) | 8.8 | - | $0 |
| 416 | Gemma 4 E4B (Non-reasoning) | 8.7 | - | $0.04 |
| 417 | NVIDIA Nemotron Nano 9B V2 (Reasoning) | 8.7 | - | $0.07 |
| 418 | Granite 4.1 30B | 8.7 | 10.4 | $0 |
| 419 | NVIDIA Nemotron 3 Nano 4B | 8.6 | 8 | $0 |
| 420 | Hermes 4 - Llama-3.1 405B (Non-reasoning) | 8.6 | - | $1.5 |
| 421 | Gemini 2.0 Flash-Lite (Feb '25) | 8.6 | - | $0 |
| 422 | Llama Nemotron Super 49B v1.5 (Non-reasoning) | 8.5 | - | $0.4 |
| 423 | Qwen3 32B (Non-reasoning) | 8.5 | - | $1.225 |
| 424 | Kimi Linear 48B A3B Instruct | 8.4 | - | $0 |
| 425 | K2-V2 (low) | 8.4 | - | $0 |
| 426 | GPT-4o (May '24) | 8.4 | 24.2 | $7.5 |
| 427 | Gemini 2.0 Flash-Lite (Preview) | 8.4 | - | $0 |
| 428 | Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning) | 8.4 | - | $0 |
| 429 | Llama 3.1 Instruct 405B | 8.3 | - | $0 |
| 430 | Llama 3.3 Nemotron Super 49B v1 (Non-reasoning) | 8.3 | - | $0 |
| 431 | Qwen3 8B (Reasoning) | 8.3 | 9 | $0.66 |
| 432 | Qwen3 VL 8B Instruct | 8.2 | - | $0.31 |
| 433 | Qwen3 4B (Reasoning) | 8.2 | - | $0 |
| 434 | LFM2.5-8B-A1B | 8.1 | - | $0 |
| 435 | GPT-4o (ChatGPT) | 8.1 | - | $0 |
| 436 | Claude 3.5 Sonnet (June '24) | 8.1 | 26 | $6 |
| 437 | Llama 3.1 Tulu3 405B | 8.1 | - | $0 |
| 438 | Ring-flash-2.0 | 8 | - | $0.247 |
| 439 | Pixtral Large | 8 | - | $0 |
| 440 | Olmo 3.1 32B Think | 7.9 | - | $0 |
| 441 | GPT-5 nano (minimal) | 7.8 | - | $0.138 |
| 442 | Gemini 1.5 Flash (Sep '24) | 7.8 | - | $0 |
| 443 | Grok 2 (Dec '24) | 7.8 | - | $0 |
| 444 | GPT-4 Turbo | 7.7 | 21.5 | $15 |
| 445 | Qwen3 VL 4B (Reasoning) | 7.7 | - | $0 |
| 446 | Solar Pro 2 (Non-reasoning) | 7.6 | - | $0 |
| 447 | Command A | 7.5 | - | $4.375 |
| 448 | Nova Pro | 7.5 | - | $1.4 |
| 449 | Llama 3.1 Nemotron Instruct 70B | 7.4 | - | $1.2 |
| 450 | Qwen3.5 2B (Reasoning) | 7.4 | 2.9 | $0 |
| 451 | Llama 3.1 Instruct 8B | 7.4 | 5.4 | $0.028 |
| 452 | Gemma 3 27B Instruct | 7.4 | 10.1 | $0 |
| 453 | Grok Beta | 7.3 | - | $0 |
| 454 | NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning) | 7.2 | - | $0.088 |
| 455 | NVIDIA Nemotron Nano 9B V2 (Non-reasoning) | 7.2 | - | $0.086 |
| 456 | Qwen2.5 Instruct 32B | 7.2 | - | $0 |
| 457 | Ministral 3 3B | 7.1 | 4.8 | $0.1 |
| 458 | Mistral Large 2 (Jul '24) | 7 | - | $3 |
| 459 | Qwen2.5 Coder Instruct 32B | 6.9 | - | $0 |
| 460 | Qwen3 4B 2507 Instruct | 6.9 | - | $0 |
| 461 | GPT-4 | 6.8 | 13.1 | $37.5 |
| 462 | GLM-4.5V (Non-reasoning) | 6.8 | - | $0.9 |
| 463 | Qwen3 14B (Non-reasoning) | 6.8 | - | $0.612 |
| 464 | Hermes 4 - Llama-3.1 70B (Non-reasoning) | 6.7 | - | $0.198 |
| 465 | GPT-4o mini | 6.7 | 11.4 | $0.262 |
| 466 | Gemini 2.5 Flash-Lite (Non-reasoning) | 6.7 | - | $0.175 |
| 467 | Mistral Small 3 | 6.7 | - | $0.15 |
| 468 | Nova Lite | 6.7 | - | $0.105 |
| 469 | Qwen3 30B A3B (Non-reasoning) | 6.6 | - | $0.35 |
| 470 | Llama 3.1 Instruct 70B | 6.5 | - | $0.56 |
| 471 | DeepSeek-V2.5 (Dec '24) | 6.5 | - | $0 |
| 472 | Qwen3 4B (Non-reasoning) | 6.5 | - | $0 |
| 473 | Granite 4.1 8B | 6.4 | 9.5 | $0.063 |
| 474 | Sarvam 30B (high) | 6.4 | - | $0.047 |
| 475 | Gemini 2.0 Flash Thinking Experimental (Dec '24) | 6.4 | - | $0 |
| 476 | DeepSeek-V2.5 | 6.4 | - | $0 |
| 477 | Gemma 4 E2B (Non-reasoning) | 6.2 | - | $0 |
| 478 | Olmo 3.1 32B Instruct | 6.2 | - | $0 |
| 479 | Mistral Saba | 6.2 | - | $0 |
| 480 | DeepSeek R1 Distill Llama 8B | 6.2 | - | $0 |
| 481 | Gemini 1.5 Pro (May '24) | 6.1 | 19.8 | $0 |
| 482 | Olmo 3 32B Think | 6.1 | - | $0 |
| 483 | Llama 3.2 Instruct 90B (Vision) | 6 | - | $0 |
| 484 | R1 1776 | 6 | - | $0 |
| 485 | Solar Mini | 6 | - | $0.15 |
| 486 | Reka Flash (Sep '24) | 6 | - | $0.35 |
| 487 | Qwen2.5 Turbo | 6 | - | $0.088 |
| 488 | Grok-1 | 5.8 | - | $0 |
| 489 | Phi-4 Mini Instruct | 5.7 | 3.8 | $0 |
| 490 | EXAONE 4.0 32B (Non-reasoning) | 5.7 | - | $0 |
| 491 | Qwen2 Instruct 72B | 5.7 | - | $0 |
| 492 | Gemma 3 12B Instruct | 5.5 | 5.8 | $0 |
| 493 | Qwen3.5 2B (Non-reasoning) | 5.3 | 2.4 | $0 |
| 494 | Qwen3.5 0.8B (Reasoning) | 5.2 | 0 | $0 |
| 495 | Gemini 1.5 Flash-8B | 5.2 | - | $0 |
| 496 | DeepHermes 3 - Mistral 24B Preview (Non-reasoning) | 5 | - | $0 |
| 497 | Jamba 1.7 Large | 5 | - | $0 |
| 498 | Granite 4.0 H Small | 4.9 | - | $0.107 |
| 499 | Qwen3 Omni 30B A3B Instruct | 4.8 | - | $0.43 |
| 500 | Hermes 3 - Llama-3.1 70B | 4.8 | - | $0.7 |
| 501 | Jamba 1.5 Large | 4.8 | - | $3.5 |
| 502 | Qwen3 8B (Non-reasoning) | 4.8 | - | $0.31 |
| 503 | DeepSeek-Coder-V2 | 4.7 | - | $0 |
| 504 | OLMo 2 32B | 4.7 | - | $0 |
| 505 | Jamba 1.6 Large | 4.7 | - | $0 |
| 506 | Phi-4 | 4.6 | - | $0.219 |
| 507 | LFM2 24B A2B | 4.6 | - | $0 |
| 508 | Gemini 1.5 Flash (May '24) | 4.6 | - | $0 |
| 509 | Nova Micro | 4.4 | - | $0.061 |
| 510 | Granite 4.1 3B | 4.4 | 4.7 | $0 |
| 511 | Claude 3 Sonnet | 4.4 | - | $0 |
| 512 | Gemini 1.0 Ultra | 4.3 | 17.6 | $0 |
| 513 | Mistral Small (Sep '24) | 4.3 | - | $0.3 |
| 514 | Phi-3 Mini Instruct 3.8B | 4.3 | - | $0 |
| 515 | Phi-4 Multimodal Instruct | 4.2 | - | $0 |
| 516 | NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning) | 4.2 | - | $0.3 |
| 517 | Gemma 3n E4B Instruct Preview (May '25) | 4.2 | - | $0 |
| 518 | Mistral Large (Feb '24) | 4.1 | - | $6 |
| 519 | Qwen2.5 Coder Instruct 7B | 4.1 | - | $0 |
| 520 | Mixtral 8x22B Instruct | 4 | - | $0 |
| 521 | Llama 3.2 Instruct 3B | 3.9 | - | $0 |
| 522 | Llama 2 Chat 7B | 3.9 | - | $0.1 |
| 523 | MiniCPM-V 4.6 1.3B | 3.8 | 0.7 | $0 |
| 524 | Jamba Reasoning 3B | 3.8 | - | $0 |
| 525 | Reka Flash 3 | 3.7 | - | $0.35 |
| 526 | Qwen1.5 Chat 110B | 3.7 | - | $0 |
| 527 | Qwen3 VL 4B Instruct | 3.7 | - | $0 |
| 528 | Olmo 3 7B Think | 3.6 | - | $0 |
| 529 | Claude 3 Haiku | 3.5 | - | $0.5 |
| 530 | Claude 2.1 | 3.5 | 14 | $0 |
| 531 | OLMo 2 7B | 3.5 | - | $0 |
| 532 | Molmo 7B-D | 3.4 | - | $0 |
| 533 | Ling-mini-2.0 | 3.4 | - | $0 |
| 534 | Claude 2.0 | 3.3 | 12.9 | $0 |
| 535 | DeepSeek R1 Distill Qwen 1.5B | 3.3 | - | $0 |
| 536 | DeepSeek-V2-Chat | 3.3 | - | $0 |
| 537 | GPT-3.5 Turbo | 3.2 | 10.7 | $0.75 |
| 538 | Mistral Small (Feb '24) | 3.2 | - | $0.262 |
| 539 | Mistral Medium | 3.2 | - | $3 |
| 540 | Llama 3 Instruct 70B | 3.1 | - | $1.175 |
| 541 | Llama 3.2 Instruct 11B (Vision) | 3 | - | $0.345 |
| 542 | LFM 40B | 3 | - | $0 |
| 543 | Arctic Instruct | 3 | - | $0 |
| 544 | Qwen Chat 72B | 3 | - | $0 |
| 545 | Qwen3.5 0.8B (Non-reasoning) | 2.9 | 1.2 | $0 |
| 546 | PALM-2 | 2.8 | 4.6 | $0 |
| 547 | Gemini 1.0 Pro | 2.7 | - | $0 |
| 548 | DeepSeek Coder V2 Lite Instruct | 2.7 | - | $0 |
| 549 | Llama 2 Chat 13B | 2.6 | - | $0 |
| 550 | Llama 2 Chat 70B | 2.6 | - | $0 |
| 551 | DeepSeek LLM 67B Chat (V1) | 2.6 | - | $0 |
| 552 | OpenChat 3.5 (1210) | 2.6 | - | $0 |
| 553 | DBRX Instruct | 2.6 | - | $0 |
| 554 | Sarvam M (Reasoning) | 2.6 | - | $0 |
| 555 | Command-R+ (Apr '24) | 2.6 | - | $6 |
| 556 | Exaone 4.0 1.2B (Reasoning) | 2.5 | - | $0 |
| 557 | Olmo 3 7B Instruct | 2.4 | - | $0.125 |
| 558 | Exaone 4.0 1.2B (Non-reasoning) | 2.4 | - | $0 |
| 559 | LFM2.5-1.2B-Thinking | 2.3 | - | $0 |
| 560 | LFM2 2.6B | 2.3 | - | $0 |
| 561 | LFM2.5-1.2B-Instruct | 2.3 | - | $0 |
| 562 | Jamba 1.7 Mini | 2.3 | - | $0 |
| 563 | Jamba 1.5 Mini | 2.3 | - | $0.25 |
| 564 | Granite 4.0 H 1B | 2.2 | - | $0 |
| 565 | Qwen3 1.7B (Reasoning) | 2.2 | - | $0 |
| 566 | Jamba 1.6 Mini | 2.1 | - | $0 |
| 567 | Gemma 3 270M | 2 | - | $0 |
| 568 | Granite 4.0 Micro | 2 | - | $0 |
| 569 | Apertus 70B Instruct | 2 | - | $1.345 |
| 570 | Mixtral 8x7B Instruct | 2 | - | $0.512 |
| 571 | DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning) | 1.9 | - | $0 |
| 572 | Llama 65B | 1.7 | - | $0 |
| 573 | Qwen Chat 14B | 1.7 | - | $0 |
| 574 | Claude Instant | 1.7 | 7.8 | $0 |
| 575 | Mistral 7B Instruct | 1.7 | - | $0.25 |
| 576 | Command-R (Mar '24) | 1.7 | - | $0.75 |
| 577 | Molmo2-8B | 1.6 | - | $0 |
| 578 | Granite 4.0 1B | 1.6 | - | $0 |
| 579 | LFM2 8B A1B | 1.3 | - | $0 |
| 580 | Granite 3.3 8B (Non-reasoning) | 1.3 | - | $0.085 |
| 581 | Qwen3 1.7B (Non-reasoning) | 1.1 | - | $0 |
| 582 | LFM2.5-VL-1.6B | 1 | - | $0 |
| 583 | Granite 4.0 H 350M | 1 | - | $0 |
| 584 | Granite 4.0 350M | 1 | - | $0 |
| 585 | Apertus 8B Instruct | 1 | - | $0.125 |
| 586 | Tiny Aya Global | 1 | - | $0 |
| 587 | Llama 3 Instruct 8B | 1 | - | $0.07 |
| 588 | Llama 3.2 Instruct 1B | 1 | - | $0 |
| 589 | Gemma 3n E4B Instruct | 1 | 3.2 | $0.075 |
| 590 | Gemma 3 4B Instruct | 1 | 2.7 | $0 |
| 591 | Gemma 3 1B Instruct | 1 | - | $0 |
| 592 | Gemma 3n E2B Instruct | 1 | - | $0 |
| 593 | LFM2 1.2B | 1 | - | $0 |
| 594 | Qwen3 0.6B (Reasoning) | 1 | - | $0 |
| 595 | Qwen3 0.6B (Non-reasoning) | 1 | - | $0 |
| 596 | GPT-5.5 Pro (xhigh) | - | - | $0 |
| 597 | Gemini 3 Deep Think | - | - | $0 |
| 598 | Claude Sonnet 5 (Adaptive Reasoning, Low Effort) | - | - | $4 |
| 599 | Claude Sonnet 5 (Adaptive Reasoning, Xhigh Effort) | - | - | $4 |
| 600 | Claude Sonnet 5 (Adaptive Reasoning, Medium Effort) | - | - | $4 |
| 601 | Claude Sonnet 5 (Adaptive Reasoning, High Effort) | - | - | $4 |
| 602 | EXAONE 4.5 33B (Non-reasoning) | - | - | $0 |
| 603 | Cogito v2.1 (Reasoning) | - | - | $1.25 |
| 604 | GPT-4o mini Realtime (Dec '24) | - | - | $0 |
| 605 | GPT-5.4 Pro (xhigh) | - | - | $67.5 |
| 606 | GPT-3.5 Turbo (0613) | - | - | $0 |
| 607 | GPT-4o Realtime (Dec '24) | - | - | $0 |
| 608 | Mi:dm K 2.5 Pro Preview | - | - | $0 |
榜单解读建议
参考 AI 大模型排行榜 时，应综合考虑“综合指数”与“成本价格”。如果您是开发者，编程能力 (Coding) 是更核心的指标。
值品工具箱同步的 AI 大模型排行榜 数据每 24 小时更新，确保您获取到最新的模型性能对比。
指标说明
- ● 综合指数：评估通用理解与逻辑。
- ● 价格 $/1M：混合 3:1 输入输出比的平均成本。
- ● 编程能力：衡量代码生成的准确性。
AI 大模型排行榜 常见问题 (FAQ)
Q1: AI 大模型排行榜 的数据多久更新？
AI 大模型排行榜 数据每 24 小时自动抓取一次，确保最新模型加入列表。
Q2: 这个 AI 大模型排行榜 包含国产模型吗？
是的，只要国产模型通过了 Artificial Analysis 的全球测评，就会出现在 AI 大模型排行榜 中。
Q3: 综合指数在 AI 大模型排行榜 中代表什么？
它代表模型的全能表现。AI 大模型排行榜 通过加权算法给出这个综合评分。
Q4: 如何在 AI 大模型排行榜 中查找性价比最高的游戏？
在 AI 大模型排行榜 页面中，您可以点击“价格”标题进行排序，寻找低价高分的模型。
Q5: AI 大模型排行榜 的编程能力测试准吗？
AI 大模型排行榜 参考了 LiveCodeBench 等权威基准测试，具有极高的参考价值。
Q6: 为什么有的新模型没进入 AI 大模型排行榜？
模型进入 AI 大模型排行榜 需要经过一系列测试，通常在新模型发布后数日内会完成更新。
Q7: AI 大模型排行榜 中的价格计算标准是什么？
价格是基于百万 Token 的调用成本，由 AI 大模型排行榜 统一混合计算得出。
Q8: 手机上能查看 AI 大模型排行榜 吗？
当然可以。AI 大模型排行榜 进行了移动端响应式深度优化。
Q9: AI 大模型排行榜 这个工具免费吗？
是的，由值品工具箱免费提供 AI 大模型排行榜 信息查询服务。
Q10: 我该怎么利用 AI 大模型排行榜 做选型？
如果您需要智能客服，参考 AI 大模型排行榜 的综合指数；如果做翻译，参考编程外的语言指标。