Paragraph Counting
Subcategory of Utility. 200 models scored.
Model Leaderboard
All models ranked by their Paragraph Counting subcategory score.
| # | Model | Paragraph Counting | Utility | Overall |
|---|---|---|---|---|
| 1 | Claude Opus 5.5 (Reasoning) | 100.00% | 98.84% | 96.43% |
| 2 | Qwen 3.8 Max (Reasoning, XHigh) | 100.00% | 100.00% | 96.21% |
| 3 | GPT-6.1 Sol (Reasoning, Medium) | 100.00% | 99.97% | 95.88% |
| 4 | Gemini 3.8 Flash (Reasoning, Medium) | 100.00% | 99.20% | 95.72% |
| 5 | Z.AI GLM 5.3 (Reasoning, Max) | 100.00% | 98.44% | 95.20% |
| 6 | GPT-5.6 Sol (Reasoning, Medium) | 100.00% | 97.51% | 95.15% |
| 7 | Claude Opus 4.6 (Reasoning) | 100.00% | 98.93% | 95.06% |
| 8 | Gemini 3.6 Flash (Reasoning) | 100.00% | 98.56% | 94.90% |
| 9 | Gemini 3.7 Flash (Reasoning, Medium) | 100.00% | 97.86% | 94.85% |
| 10 | Hy4 Preview (Reasoning, High) | 100.00% | 99.90% | 94.80% |
| 11 | Muse Spark 1.2 (Reasoning, Medium) | 100.00% | 99.95% | 94.71% |
| 12 | Claude Haiku 5.5 (Reasoning, Max) | 100.00% | 99.33% | 94.65% |
| 13 | Qwen 3.7 Max | 100.00% | 99.54% | 94.55% |
| 14 | DeepSeek V4.1 Flash (Reasoning, High) | 100.00% | 99.43% | 94.30% |
| 15 | Z.AI GLM 5.3 Flash (Reasoning, Max) | 100.00% | 98.50% | 94.28% |
| 16 | Qwen 3.8 Flash (Reasoning) | 100.00% | 99.39% | 94.24% |
| 17 | Gemini 3.1 Pro (Preview) | 100.00% | 99.91% | 94.08% |
| 18 | Muse Spark 1.3 (Reasoning, Medium) | 100.00% | 98.62% | 94.06% |
| 19 | GPT-6 Luna (Reasoning, High) | 100.00% | 92.68% | 93.95% |
| 20 | GPT-5.4 (Reasoning) | 100.00% | 96.89% | 93.85% |
| 21 | Muse Spark 1.1 (Reasoning, Medium) | 100.00% | 99.40% | 93.81% |
| 22 | Z.AI GLM 5.1 | 100.00% | 97.51% | 93.74% |
| 23 | Qwen 3.6 Max Preview | 100.00% | 98.34% | 93.72% |
| 24 | GPT-5.5 (Reasoning, Medium) | 100.00% | 96.60% | 93.72% |
| 25 | Claude Sonnet 4.6 (Reasoning) | 100.00% | 97.88% | 93.64% |
| 26 | Z.AI GLM 5.2 (Reasoning, High) | 100.00% | 96.31% | 93.41% |
| 27 | Gemini 3.5 Flash (Reasoning) | 100.00% | 98.86% | 93.35% |
| 28 | Z.AI GLM 5 Turbo | 100.00% | 96.36% | 93.29% |
| 29 | Kimi K3 (Reasoning, High) | 100.00% | 96.19% | 92.99% |
| 30 | Qwen 3.8 27B (Reasoning, XHigh) | 100.00% | 99.36% | 92.91% |
| 31 | Claude Sonnet 5.5 (Reasoning) | 100.00% | 94.61% | 92.77% |
| 32 | Grok 4.6 (Reasoning, High) | 100.00% | 83.88% | 92.76% |
| 33 | GPT-6 Sol (Reasoning, Medium) | 100.00% | 91.07% | 92.63% |
| 34 | Grok 4.7 (Reasoning, High) | 100.00% | 91.12% | 92.62% |
| 35 | Claude Opus 4.7 (Reasoning) | 100.00% | 97.87% | 92.53% |
| 36 | GPT-5.5 (Reasoning, Low) | 100.00% | 96.36% | 92.51% |
| 37 | GPT-5.6 Terra (Reasoning, Medium) | 100.00% | 95.58% | 92.49% |
| 38 | GPT-6 Luna (Reasoning, Medium) | 100.00% | 91.17% | 92.40% |
| 39 | Aion 3.5 Mini (Reasoning, High) | 100.00% | 95.21% | 92.36% |
| 40 | Claude Opus 4.8 (Reasoning) | 100.00% | 99.26% | 92.33% |
| 41 | Claude Opus 4.6 | 100.00% | 90.72% | 92.31% |
| 42 | Mistral Large 4.0 (Reasoning) | 100.00% | 97.35% | 92.14% |
| 43 | Kimi K3 (Reasoning, Low) | 100.00% | 94.04% | 92.07% |
| 44 | Claude Opus 5 (Reasoning, Low) | 100.00% | 92.75% | 92.06% |
| 45 | Claude Opus 4.8 (Reasoning, Low) | 100.00% | 98.00% | 91.89% |
| 46 | Xiaomi MIMO v2.6 Pro | 100.00% | 95.96% | 91.50% |
| 47 | GPT-5 | 100.00% | 93.53% | 91.48% |
| 48 | Muse Spark 1.1 (Reasoning, Minimal) | 100.00% | 98.38% | 91.43% |
| 49 | GPT-5 Mini | 100.00% | 98.39% | 91.31% |
| 50 | Claude Opus 5 (Reasoning) | 100.00% | 93.39% | 91.15% |
| 51 | Qwen 3.5 397B A17B | 100.00% | 97.50% | 91.09% |
| 52 | Grok 4.3 (Reasoning) | 100.00% | 92.94% | 90.99% |
| 53 | GPT-5.4 (Reasoning, Low) | 100.00% | 95.32% | 90.91% |
| 54 | Grok 4.20 (Reasoning) | 100.00% | 92.61% | 90.87% |
| 55 | Kimi K2.5 | 100.00% | 96.63% | 90.86% |
| 56 | GPT-5.6 Sol | 100.00% | 80.44% | 90.76% |
| 57 | GPT-5.1 | 100.00% | 95.33% | 90.73% |
| 58 | Claude Sonnet 4.6 | 100.00% | 88.52% | 90.66% |
| 59 | GPT-5.6 Luna (Reasoning, Medium) | 100.00% | 92.18% | 90.61% |
| 60 | MiniMax M3 | 100.00% | 93.59% | 90.45% |
| 61 | Claude Sonnet 5 (Reasoning) | 100.00% | 92.87% | 90.40% |
| 62 | Qwen 3.5 122B | 100.00% | 96.36% | 90.32% |
| 63 | DeepSeek V4 Pro 0813 (Reasoning, High) | 100.00% | 97.90% | 90.21% |
| 64 | Claude Sonnet 5 (Reasoning, Low) | 100.00% | 93.53% | 90.16% |
| 65 | GPT-6 Sol | 100.00% | 81.18% | 90.05% |
| 66 | Qwen 3.5 27B | 100.00% | 95.67% | 90.05% |
| 67 | Claude Opus 5 | 100.00% | 90.50% | 89.97% |
| 68 | Gemini 3 Flash (Preview, Reasoning) | 100.00% | 97.20% | 89.93% |
| 69 | Claude Opus 4.7 | 100.00% | 95.77% | 89.90% |
| 70 | GPT-5.4 Mini (Reasoning) | 100.00% | 94.44% | 89.82% |
| 71 | Qwen 3.5 Plus (2026-04-20) | 100.00% | 96.42% | 89.79% |
| 72 | Gemma 4 31B (Reasoning) | 100.00% | 96.32% | 89.64% |
| 73 | Claude Opus 4.5 | 100.00% | 89.84% | 89.60% |
| 74 | Z.AI GLM 5 | 100.00% | 94.11% | 89.60% |
| 75 | ByteDance Seed 1.6 | 100.00% | 90.83% | 89.59% |
| 76 | GPT-5.2 | 100.00% | 96.22% | 89.45% |
| 77 | GPT-5.5 | 100.00% | 81.88% | 89.37% |
| 78 | Qwen 3.6 Flash | 100.00% | 96.09% | 89.31% |
| 79 | DeepSeek V4 Pro (Reasoning) | 100.00% | 93.24% | 89.28% |
| 80 | Gemma 4 26B (Reasoning) | 100.00% | 95.69% | 89.02% |
| 81 | Muse Glimmer 30B (Reasoning, Medium) | 100.00% | 96.62% | 88.81% |
| 82 | Aion 3.0 | 100.00% | 93.57% | 88.78% |
| 83 | o4 Mini High | 100.00% | 98.67% | 88.78% |
| 84 | Z.AI GLM 5.3 Flash (Reasoning, Low) | 100.00% | 90.95% | 88.76% |
| 85 | Qwen 3.7 Flash (Reasoning) | 100.00% | 96.06% | 88.56% |
| 86 | Gemini 2.5 Pro | 100.00% | 92.18% | 88.44% |
| 87 | Qwen 3.6 27B | 100.00% | 94.32% | 88.33% |
| 88 | GPT-5.6 Terra | 100.00% | 77.09% | 88.23% |
| 89 | DeepSeek V4 Flash (Reasoning) | 100.00% | 87.53% | 88.06% |
| 90 | DeepSeek V4 Flash 0731 (Reasoning, High) | 100.00% | 94.25% | 87.97% |
| 91 | Z.AI GLM 4.7 | 100.00% | 94.31% | 87.67% |
| 92 | Qwen 3.6 35B | 100.00% | 96.20% | 87.66% |
| 93 | Z.AI GLM 4.6 | 100.00% | 88.58% | 87.64% |
| 94 | Claude Sonnet 4 | 100.00% | 84.02% | 87.64% |
| 95 | Claude Sonnet 4.5 | 100.00% | 83.78% | 87.54% |
| 96 | Claude Sonnet 5 | 100.00% | 88.57% | 87.34% |
| 97 | Xiaomi MIMO v2.6 Flash | 100.00% | 88.05% | 87.31% |
| 98 | Claude Opus 4 | 100.00% | 88.81% | 87.22% |
| 99 | Qwen 3.5 35B | 100.00% | 96.42% | 87.01% |
| 100 | GPT-4.1 | 100.00% | 90.57% | 86.82% |
| 101 | MiniMax M2.5 | 100.00% | 90.42% | 86.71% |
| 102 | Gemini 3.6 Flash (Reasoning, Minimal) | 100.00% | 84.77% | 86.68% |
| 103 | Aion 2.0 | 100.00% | 90.91% | 86.66% |
| 104 | o4 Mini | 100.00% | 96.31% | 86.56% |
| 105 | MiniMax M2.7 | 100.00% | 95.50% | 86.23% |
| 106 | Qwen 3.5 Plus (2026-02-15) | 100.00% | 86.65% | 86.17% |
| 107 | Xiaomi MIMO v2.5 Pro | 100.00% | 82.62% | 86.05% |
| 108 | Gemini 3.1 Flash Lite (Reasoning) | 100.00% | 92.32% | 85.91% |
| 109 | Gemini 3.5 Flash (Reasoning, Minimal) | 100.00% | 83.90% | 85.88% |
| 110 | ByteDance Seed 2.0 Mini | 100.00% | 91.88% | 85.69% |
| 111 | Qwen 3.5 Flash | 100.00% | 96.11% | 85.66% |
| 112 | Gemini 3 Flash (Preview) | 100.00% | 86.39% | 85.47% |
| 113 | Gemini 3.1 Flash Lite (Preview) | 100.00% | 94.00% | 85.41% |
| 114 | Mistral Medium 3.5 (Reasoning) | 100.00% | 74.19% | 85.32% |
| 115 | Gemma 4 31B | 100.00% | 86.69% | 85.23% |
| 116 | Gemini 3.1 Flash Lite | 100.00% | 92.77% | 85.09% |
| 117 | GPT-5.6 Luna | 100.00% | 75.00% | 85.06% |
| 118 | Z.AI GLM 4.5 | 100.00% | 79.19% | 84.95% |
| 119 | Gemma 4 26B | 100.00% | 83.17% | 84.89% |
| 120 | GPT-OSS 120B | 100.00% | 92.03% | 84.81% |
| 121 | Claude Haiku 5.5 (Reasoning, Low) | 100.00% | 89.31% | 84.75% |
| 122 | Mistral Large 4.0 | 100.00% | 84.57% | 84.72% |
| 123 | Gemini 3.5 Flash Lite (Reasoning, Minimal) | 100.00% | 84.58% | 84.45% |
| 124 | Gemini 3.5 Flash Lite (Reasoning) | 100.00% | 85.46% | 84.34% |
| 125 | GPT-5.4 | 100.00% | 81.95% | 84.31% |
| 126 | Mistral Large 3 | 100.00% | 84.91% | 84.29% |
| 127 | ByteDance Seed 2.0 Lite | 100.00% | 92.23% | 84.27% |
| 128 | Gemini 2.5 Flash (Reasoning) | 100.00% | 82.25% | 84.14% |
| 129 | DeepSeek-V2 Chat | 100.00% | 83.82% | 84.09% |
| 130 | Qwen 3.5 9B | 100.00% | 94.02% | 84.05% |
| 131 | GPT-6 Luna | 100.00% | 69.00% | 83.73% |
| 132 | Aion 3.0 Mini | 100.00% | 90.76% | 83.69% |
| 133 | GPT-5.4 Mini (Reasoning, Low) | 100.00% | 88.49% | 83.57% |
| 134 | Gemini 2.5 Flash Lite (Reasoning) | 100.00% | 89.63% | 83.10% |
| 135 | DeepSeek V3 (2024-12-26) | 100.00% | 81.87% | 82.62% |
| 136 | DeepSeek V3.2 | 100.00% | 81.58% | 82.22% |
| 137 | Z.AI GLM 4.7 Flash | 100.00% | 88.98% | 82.21% |
| 138 | GPT-4o, Aug. 6th (temp=0) | 100.00% | 82.11% | 82.18% |
| 139 | DeepSeek V4 Pro | 100.00% | 77.57% | 82.05% |
| 140 | DeepSeek V4 Flash | 100.00% | 83.26% | 82.02% |
| 141 | Inception Mercury 2 | 100.00% | 92.86% | 81.99% |
| 142 | Nemotron 3 Super | 100.00% | 95.29% | 81.69% |
| 143 | Mistral Large 2 | 100.00% | 69.19% | 81.50% |
| 144 | GPT-4.1 Mini | 100.00% | 82.30% | 81.40% |
| 145 | GPT-4o, Aug. 6th (temp=1) | 100.00% | 82.44% | 81.28% |
| 146 | Grok 4.20 | 100.00% | 84.11% | 81.21% |
| 147 | Z.AI GLM 4.5 Air | 100.00% | 76.57% | 80.74% |
| 148 | Gemini 2.5 Flash | 100.00% | 61.45% | 80.61% |
| 149 | GPT-5.4 Mini | 100.00% | 79.37% | 80.45% |
| 150 | GPT-5 Nano | 100.00% | 93.91% | 80.16% |
| 151 | GPT-5.4 Nano (Reasoning) | 100.00% | 93.34% | 80.02% |
| 152 | Laguna S 2.1 | 100.00% | 81.67% | 79.99% |
| 153 | Gemini 2.5 Flash Lite | 100.00% | 80.14% | 79.91% |
| 154 | Mistral Small 4 (Reasoning) | 100.00% | 85.61% | 79.48% |
| 155 | Qwen 3 32B | 100.00% | 81.66% | 79.37% |
| 156 | Writer: Palmyra X5 | 100.00% | 79.71% | 78.11% |
| 157 | Qwen 3 235B A22B Instruct 2507 | 100.00% | 83.15% | 78.07% |
| 158 | GPT-4o Mini (temp=1) | 100.00% | 82.16% | 77.82% |
| 159 | GPT-5.4 Nano (Reasoning, Low) | 100.00% | 91.42% | 77.46% |
| 160 | Llama 3.1 70B | 100.00% | 81.03% | 77.41% |
| 161 | Mistral Small 3.2 24B | 100.00% | 73.17% | 77.36% |
| 162 | GPT-4o Mini (temp=0) | 100.00% | 81.43% | 76.86% |
| 163 | Mistral Medium 3.1 | 100.00% | 80.13% | 76.08% |
| 164 | Gemma 3 12B | 100.00% | 79.28% | 76.07% |
| 165 | Gemma 3 27B | 100.00% | 76.82% | 75.70% |
| 166 | Laguna XS 2.1 | 100.00% | 78.98% | 75.42% |
| 167 | Mistral Small 4 | 100.00% | 78.28% | 75.23% |
| 168 | Nemotron 3 Nano | 100.00% | 86.00% | 74.50% |
| 169 | Qwen 2.5 72B | 100.00% | 76.43% | 73.17% |
| 170 | GPT-5.4 Nano | 100.00% | 78.57% | 72.16% |
| 171 | ByteDance Seed 1.6 Flash | 100.00% | 84.16% | 70.92% |
| 172 | Ministral 3 14B | 100.00% | 79.03% | 70.45% |
| 173 | Ministral 3 8B | 100.00% | 74.43% | 69.98% |
| 174 | GPT-4.1 Nano | 100.00% | 68.45% | 69.90% |
| 175 | Ministral 3 3B | 100.00% | 72.38% | 65.02% |
| 176 | Grok 4.5 (Reasoning, High) | 96.67% | 86.69% | 94.12% |
| 177 | Kimi K2.6 | 96.67% | 97.42% | 92.57% |
| 178 | Aion 3.5 (Reasoning, High) | 96.67% | 95.78% | 91.90% |
| 179 | Thinking Machines Inkling (Reasoning) | 96.67% | 93.19% | 90.78% |
| 180 | Thinking Machines Inkling Small (Reasoning, High) | 96.67% | 92.99% | 87.39% |
| 181 | Mistral Medium 3.5 | 96.67% | 72.28% | 84.52% |
| 182 | Claude Haiku 5.5 (Reasoning, Medium) | 96.67% | 88.53% | 84.17% |
| 183 | Xiaomi MIMO v2.5 | 93.33% | 81.15% | 83.95% |
| 184 | DeepSeek V3.1 | 93.33% | 76.65% | 82.35% |
| 185 | DeepSeek V3 (2025-03-24) | 93.33% | 80.62% | 79.93% |
| 186 | Cydonia 24B V4.1 | 93.33% | 69.32% | 72.68% |
| 187 | Grok 4.5 (Reasoning, Low) | 90.00% | 82.63% | 90.94% |
| 188 | Claude Haiku 5.5 (Adaptive) | 90.00% | 91.83% | 89.37% |
| 189 | WizardLM 2 8x22b | 90.00% | 67.14% | 71.45% |
| 190 | Cohere Command R+ (Aug. 2024) | 90.00% | 59.51% | 67.04% |
| 191 | Hermes 3 405B | 86.67% | 69.02% | 80.80% |
| 192 | Grok 4.3 | 86.67% | 66.41% | 78.00% |
| 193 | Arcee AI: Trinity Mini | 80.00% | 59.94% | 67.68% |
| 194 | Claude Haiku 4.5 | 70.00% | 72.48% | 83.36% |
| 195 | Thinking Machines Inkling | 66.67% | 65.97% | 73.96% |
| 196 | Hermes 3 70B | 66.67% | 61.15% | 69.74% |
| 197 | Gemma 3 4B | 53.33% | 60.30% | 66.33% |
| 198 | Mistral NeMO | 40.00% | 51.55% | 63.80% |
| 199 | Ministral 8B | 33.33% | 46.82% | 63.77% |
| 200 | Ministral 3B | 33.33% | 49.17% | 59.25% |