| 1 | GPT-5.4 Nano | 44.2% | $0.0057 | 26.3s | 15% |
| 2 | ByteDance Seed 1.6 Flash | 45.7% | $0.0013 | 27.3s | 11% |
| 3 | Gemini 2.5 Flash Lite | 43.8% | $0.0009 | 9.5s | 11% |
| 4 | Gemini 2.5 Flash Lite (Reasoning) | 44.3% | $0.0028 | 30.8s | 10% |
| 5 | Qwen 3 32B | 46.1% | $0.0015 | 54.6s | 10% |
| 6 | Gemma 3 4B | 42.9% | $0.0002 | 20.0s | 9% |
| 7 | Writer: Palmyra X5 | 44.4% | $0.011 | 22.0s | 9% |
| 8 | Laguna XS 2.1 | 48.6% | $0.0003 | 14.4s | 6% |
| 9 | Muse Spark 1.1 (Reasoning, Minimal) | 40.6% | $0.012 | 19.2s | 10% |
| 10 | Gemma 3 12B | 37.9% | $0.0004 | 41.3s | 10% |
| 11 | Gemini 3.1 Flash Lite (Preview) | 47.6% | $0.0030 | 8.4s | 6% |
| 12 | GPT-5.4 Nano (Reasoning, Low) | 39.1% | $0.0055 | 20.6s | 9% |
| 13 | Z.AI GLM 5 | 41.3% | $0.0084 | 1.2m | 10% |
| 14 | Qwen 3.5 Plus (2026-02-15) | 38.5% | $0.0060 | 31.5s | 9% |
| 15 | Qwen 3.5 397B A17B | 54.2% | $0.014 | 3.0m | 10% |
| 16 | Xiaomi MIMO v2.5 | 32.9% | $0.0054 | 31.8s | 11% |
| 17 | Cydonia 24B V4.1 | 45.0% | $0.0014 | 44.8s | 7% |
| 18 | Gemini 3.1 Flash Lite (Reasoning) | 46.6% | $0.0030 | 11.9s | 5% |
| 19 | Z.AI GLM 5 Turbo | 36.5% | $0.0081 | 33.2s | 9% |
| 20 | Grok 4.20 | 34.7% | $0.0093 | 45.7s | 10% |
| 21 | Gemma 3 27B | 39.2% | $0.0006 | 52.6s | 8% |
| 22 | GPT-5.4 Mini | 33.1% | $0.015 | 16.8s | 10% |
| 23 | Qwen3 235B A22B Instruct 2507 | 39.3% | $0.0011 | 59.2s | 8% |
| 24 | Qwen 3.6 35B | 42.3% | $0.0083 | 1.0m | 8% |
| 25 | Muse Spark 1.1 (Reasoning, Medium) | 40.3% | $0.015 | 23.9s | 7% |
| 26 | Gemini 3.1 Flash Lite | 45.0% | $0.0030 | 12.1s | 4% |
| 27 | Gemini 2.5 Flash (Reasoning) | 38.0% | $0.011 | 21.5s | 7% |
| 28 | GPT-5.4 Mini (Reasoning, Low) | 32.8% | $0.015 | 16.8s | 9% |
| 29 | GPT-5.6 Terra (Reasoning) | 29.2% | $0.044 | 35.2s | 12% |
| 30 | Claude Sonnet 4 | 45.8% | $0.032 | 43.7s | 7% |
| 31 | GPT-5.4 (Reasoning, Low) | 44.9% | $0.055 | 1.4m | 10% |
| 32 | GPT-5 Nano | 38.3% | $0.0042 | 1.4m | 8% |
| 33 | Z.AI GLM 4.7 | 36.5% | $0.010 | 1.4m | 9% |
| 34 | o4 Mini | 29.5% | $0.015 | 25.7s | 9% |
| 35 | GPT-5.4 (Reasoning) | 47.6% | $0.089 | 2.6m | 14% |
| 36 | GPT-4o, Aug. 6th (temp=1) | 45.8% | $0.018 | 24.4s | 4% |
| 37 | GPT-5.4 | 44.6% | $0.049 | 1.4m | 9% |
| 38 | GPT-5.4 Mini (Reasoning) | 32.4% | $0.022 | 28.1s | 9% |
| 39 | Claude Opus 4.6 | 34.7% | $0.078 | 1.2m | 14% |
| 40 | Qwen 3.6 Flash | 36.0% | $0.010 | 41.4s | 7% |
| 41 | Grok 4.20 (Reasoning) | 37.5% | $0.018 | 1.5m | 9% |
| 42 | Aion 2.0 | 32.1% | $0.0064 | 1.3m | 9% |
| 43 | Z.AI GLM 5.1 | 39.3% | $0.014 | 1.5m | 8% |
| 44 | GPT-5.4 Nano (Reasoning) | 30.1% | $0.0061 | 24.5s | 8% |
| 45 | GPT-5 Mini | 33.0% | $0.0100 | 57.4s | 8% |
| 46 | Cohere Command R+ (Aug. 2024) | 47.1% | $0.020 | 52.5s | 3% |
| 47 | Hermes 3 70B | 47.9% | $0.0010 | 1.2m | 3% |
| 48 | GPT-5.1 | 35.4% | $0.054 | 1.8m | 12% |
| 49 | Claude Opus 4.6 (Reasoning) | 38.8% | $0.088 | 1.4m | 11% |
| 50 | GPT-5.6 Terra | 28.8% | $0.046 | 36.9s | 9% |
| 51 | Qwen 3.5 35B | 36.3% | $0.018 | 1.0m | 5% |
| 52 | Mistral Small 4 (Reasoning) | 31.0% | $0.0022 | 30.2s | 4% |
| 53 | Mistral NeMO | 38.7% | $0.0005 | 10.1s | 0% |
| 54 | Qwen 3.5 Flash | 35.4% | $0.0025 | 47.5s | 3% |
| 55 | DeepSeek V4 Pro (Reasoning) | 38.5% | $0.015 | 3.1m | 8% |
| 56 | GPT-4o Mini (temp=1) | 40.9% | $0.0012 | 34.8s | 0% |
| 57 | Laguna S 2.1 | 38.9% | $0.0005 | 22.1s | 0% |
| 58 | GPT-5.6 Sol (Reasoning) | 29.8% | $0.108 | 1.2m | 12% |
| 59 | GPT-4.1 Nano | 36.0% | $0.0007 | 13.3s | 0% |
| 60 | GPT-5.6 Luna (Reasoning) | 25.2% | $0.017 | 18.0s | 4% |
| 61 | Gemini 3.5 Flash (Reasoning, Minimal) | 37.4% | $0.018 | 12.0s | 0% |
| 62 | Ministral 3 14B | 33.2% | $0.0007 | 11.7s | 0% |
| 63 | Claude Haiku 4.5 | 35.3% | $0.011 | 21.6s | 0% |
| 64 | GPT-5 | 34.6% | $0.065 | 2.8m | 10% |
| 65 | Arcee AI: Trinity Mini | 30.9% | $0.0003 | 9.2s | 0% |
| 66 | o4 Mini High | 26.5% | $0.025 | 47.2s | 5% |
| 67 | Gemini 2.5 Flash | 31.2% | $0.0052 | 10.6s | 0% |
| 68 | Mistral Small 4 | 30.1% | $0.0014 | 18.2s | 0% |
| 69 | Z.AI GLM 4.7 Flash | 36.3% | $0.0017 | 1.2m | 0% |
| 70 | Gemini 3 Flash (Preview, Reasoning) | 33.5% | $0.012 | 30.1s | 0% |
| 71 | GPT-5.5 (Reasoning, Low) | 34.3% | $0.139 | 1.8m | 12% |
| 72 | DeepSeek V4 Flash (Reasoning) | 30.2% | $0.0007 | 31.1s | 0% |
| 73 | Qwen 3.5 Plus (2026-04-20) | 30.6% | $0.017 | 1.8m | 4% |
| 74 | GPT-5.5 | 32.9% | $0.139 | 1.7m | 12% |
| 75 | GPT-5.2 | 24.8% | $0.056 | 1.5m | 8% |
| 76 | Claude Sonnet 4.6 | 37.6% | $0.031 | 39.3s | 0% |
| 77 | Z.AI GLM 4.5 | 31.8% | $0.0051 | 42.1s | 0% |
| 78 | Ministral 3 8B | 27.6% | $0.0008 | 19.6s | 0% |
| 79 | Gemini 3.5 Flash Lite (Reasoning, Minimal) | 26.2% | $0.0035 | 6.3s | 0% |
| 80 | Gemma 4 26B | 31.6% | $0.0009 | 55.1s | 0% |
| 81 | Z.AI GLM 4.5 Air | 32.3% | $0.0029 | 58.2s | 0% |
| 82 | Gemini 3.6 Flash (Reasoning, Minimal) | 29.9% | $0.017 | 14.3s | 0% |
| 83 | Z.AI GLM 4.6 | 32.0% | $0.0065 | 51.5s | 0% |
| 84 | GPT-5.6 Sol | 23.7% | $0.105 | 1.2m | 10% |
| 85 | Mistral Large 3 | 28.0% | $0.0033 | 30.3s | 0% |
| 86 | Mistral Medium 3.1 | 28.3% | $0.0048 | 36.5s | 0% |
| 87 | Ministral 8B | 23.9% | $0.0004 | 10.4s | 0% |
| 88 | Hermes 3 405B | 29.9% | $0.0032 | 53.2s | 0% |
| 89 | GPT-4.1 Mini | 25.4% | $0.0027 | 19.0s | 0% |
| 90 | GPT-4o Mini (temp=0) | 27.0% | $0.0012 | 34.8s | 0% |
| 91 | Gemma 4 31B | 34.7% | $0.0010 | 1.6m | 0% |
| 92 | Aion 3.0 | 35.3% | $0.024 | 1.0m | 0% |
| 93 | Gemini 3 Flash (Preview) | 25.2% | $0.0078 | 19.6s | 0% |
| 94 | Qwen3.6 Max Preview | 37.1% | $0.050 | 3.5m | 7% |
| 95 | DeepSeek V3.2 | 34.6% | $0.0014 | 1.9m | 0% |
| 96 | Gemini 3.5 Flash Lite (Reasoning) | 21.5% | $0.0035 | 6.6s | 0% |
| 97 | Mistral Large 2 | 26.5% | $0.013 | 29.4s | 0% |
| 98 | DeepSeek-V2 Chat | 26.5% | $0.0021 | 53.3s | 0% |
| 99 | MiniMax M2.7 | 28.6% | $0.0040 | 1.1m | 0% |
| 100 | DeepSeek V4 Flash | 23.2% | $0.0006 | 31.6s | 0% |
| 101 | GPT-5.5 (Reasoning) | 30.1% | $0.142 | 1.8m | 11% |
| 102 | Llama 3.1 70B | 23.0% | $0.0015 | 29.4s | 0% |
| 103 | MiniMax M2.5 | 29.0% | $0.0034 | 1.3m | 0% |
| 104 | DeepSeek V3 (2024-12-26) | 25.4% | $0.0021 | 54.6s | 0% |
| 105 | Claude Sonnet 5 | 28.2% | $0.027 | 33.5s | 0% |
| 106 | Aion 3.0 Mini | 27.6% | $0.0053 | 1.2m | 0% |
| 107 | Xiaomi MIMO v2.5 Pro | 26.3% | $0.0085 | 53.5s | 0% |
| 108 | DeepSeek V3.1 | 31.7% | $0.0020 | 1.8m | 0% |
| 109 | Ministral 3B | 18.4% | $0.0001 | 8.1s | 0% |
| 110 | Qwen 2.5 72B | 22.0% | $0.0010 | 36.7s | 0% |
| 111 | Nemotron 3 Nano | 25.5% | $0.0010 | 1.1m | 0% |
| 112 | Gemini 2.5 Pro | 29.5% | $0.036 | 36.2s | 0% |
| 113 | Gemma 4 31B (Reasoning) | 33.1% | $0.0014 | 2.2m | 0% |
| 114 | DeepSeek V4 Pro | 27.0% | $0.0048 | 1.3m | 0% |
| 115 | GPT-4.1 | 25.9% | $0.018 | 44.7s | 0% |
| 116 | GPT-5.6 Luna | 21.4% | $0.015 | 15.8s | 0% |
| 117 | Gemini 3.6 Flash (Reasoning) | 30.0% | $0.044 | 32.5s | 0% |
| 118 | Z.AI GLM 5.2 (Reasoning, High) | 25.8% | $0.011 | 1.0m | 0% |
| 119 | GPT-4o, Aug. 6th (temp=0) | 23.4% | $0.023 | 22.7s | 0% |
| 120 | Qwen 3.5 122B | 29.0% | $0.025 | 1.1m | 0% |
| 121 | ByteDance Seed 1.6 | 36.2% | $0.013 | 2.5m | 0% |
| 122 | Gemini 3.5 Flash (Reasoning) | 35.3% | $0.071 | 37.6s | 0% |
| 123 | Grok 4.3 | 19.1% | $0.0069 | 30.5s | 0% |
| 124 | Claude Sonnet 5 (Reasoning) | 25.4% | $0.030 | 38.9s | 0% |
| 125 | WizardLM 2 8x22b | 27.8% | $0.0026 | 1.8m | 0% |
| 126 | Claude Sonnet 4.5 | 26.0% | $0.035 | 38.1s | 0% |
| 127 | MoonshotAI: Kimi K3 (Reasoning, Low) | 27.8% | $0.028 | 1.1m | 0% |
| 128 | Claude Opus 4.5 | 35.1% | $0.070 | 53.4s | 0% |
| 129 | DeepSeek V3 (2025-03-24) | 17.4% | $0.0014 | 39.4s | 0% |
| 130 | ByteDance Seed 2.0 Lite | 31.3% | $0.012 | 2.2m | 0% |
| 131 | Qwen 3.5 9B | 22.3% | $0.0011 | 1.4m | 0% |
| 132 | Ministral 3 3B | 13.0% | $0.0005 | 11.1s | 0% |
| 133 | Claude Sonnet 5 (Reasoning, Low) | 22.3% | $0.031 | 38.4s | 0% |
| 134 | Nemotron 3 Super | 21.1% | $0.0000 | 1.4m | 0% |
| 135 | Qwen 3.5 27B | 27.1% | $0.020 | 1.6m | 0% |
| 136 | Gemma 4 26B (Reasoning) | 24.7% | $0.0013 | 2.0m | 0% |
| 137 | MoonshotAI: Kimi K2.5 | 37.5% | $0.019 | 3.2m | 0% |
| 138 | Inception Mercury 2 | 9.2% | $0.0032 | 7.0s | 0% |
| 139 | Claude Opus 4.7 | 27.4% | $0.069 | 30.4s | 0% |
| 140 | Claude Opus 4.8 (Reasoning, Low) | 28.3% | $0.071 | 41.9s | 0% |
| 141 | Claude Opus 5 | 31.6% | $0.082 | 50.1s | 0% |
| 142 | Thinking Machines Inkling | 18.7% | $0.0075 | 1.5m | 0% |
| 143 | Qwen 3.6 27B | 28.4% | $0.025 | 2.3m | 0% |
| 144 | Claude Sonnet 4.6 (Reasoning) | 27.8% | $0.060 | 1.2m | 0% |
| 145 | Claude Opus 4 | 39.1% | $0.209 | 1.4m | 8% |
| 146 | Claude Opus 4.7 (Reasoning) | 26.0% | $0.076 | 32.0s | 0% |
| 147 | Grok 4.5 (Reasoning, Low) | 15.7% | $0.018 | 1.1m | 0% |
| 148 | MiniMax M3 | 27.6% | $0.0060 | 3.1m | 0% |
| 149 | Claude Opus 4.8 (Reasoning) | 24.1% | $0.071 | 41.7s | 0% |
| 150 | MoonshotAI: Kimi K3 (Reasoning, High) | 26.1% | $0.044 | 1.8m | 0% |
| 151 | ByteDance Seed 2.0 Mini | 38.0% | $0.0045 | 4.9m | 0% |
| 152 | Qwen3.7 Max | 33.2% | $0.068 | 2.3m | 0% |
| 153 | GPT-OSS 120B | 11.3% | $0.0015 | 1.8m | 0% |
| 154 | Grok 4.3 (Reasoning) | 19.1% | $0.021 | 2.3m | 0% |
| 155 | Grok 4.5 (Reasoning, High) | 14.2% | $0.030 | 1.6m | 0% |
| 156 | Gemini 3.1 Pro (Preview) | 30.7% | $0.107 | 1.8m | 0% |
| 157 | Claude Opus 5 (Reasoning, Low) | 20.7% | $0.097 | 1.0m | 0% |
| 158 | Claude Opus 5 (Reasoning) | 19.4% | $0.098 | 1.0m | 0% |
| 159 | Thinking Machines Inkling (Reasoning) | 24.3% | $0.025 | 5.1m | 0% |
| 160 | MoonshotAI: Kimi K2.6 | 28.5% | $0.058 | 6.5m | 0% |
| 161 | Mistral Small 3.2 24B | 5.9% | $0.0069 | 5.7m | 0% |