Utility
32 scenarios across 5 subcategories. 118 models scored.
Subcategories
| Subcategory | Avg Score | Best Model | Best Score |
|---|---|---|---|
| Word Counting | 62.97% | Gemini 3.1 Pro (Preview) | 99.56% |
| Sentence Counting | 87.81% | Gemini 3.1 Pro (Preview) | 100.00% |
| Paragraph Counting | 95.31% | Claude Opus 4.6 (Reasoning) | 100.00% |
| Structural Counting | 72.62% | Claude Opus 4.6 (Reasoning) | 100.00% |
| Data Extraction | 97.88% | Claude Opus 4.6 (Reasoning) | 100.00% |
Model Leaderboard
All models ranked by their Utility category score.
| # | Model | Utility | Word Counting | Sentence Counting | Paragraph Counting | Structural Counting | Data Extraction | Overall |
|---|---|---|---|---|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 99.91% | 99.56% | 100.00% | 100.00% | 100.00% | 100.00% | 94.37% |
| 2 | Claude Opus 4.6 (Reasoning) | 98.93% | 94.65% | 100.00% | 100.00% | 100.00% | 100.00% | 95.02% |
| 3 | o4 Mini High | 98.67% | 95.97% | 99.90% | 100.00% | 97.50% | 100.00% | 90.29% |
| 4 | GPT-5 Mini | 98.39% | 97.93% | 100.00% | 100.00% | 94.00% | 100.00% | 92.62% |
| 5 | Claude Sonnet 4.6 (Reasoning) | 97.88% | 90.42% | 99.96% | 100.00% | 99.00% | 100.00% | 93.66% |
| 6 | Qwen 3.5 397B A17B | 97.50% | 87.52% | 100.00% | 100.00% | 100.00% | 100.00% | 91.73% |
| 7 | Gemini 3 Flash (Preview, Reasoning) | 97.20% | 86.00% | 100.00% | 100.00% | 100.00% | 100.00% | 90.50% |
| 8 | GPT-5.4 (Reasoning) | 96.89% | 84.46% | 100.00% | 100.00% | 100.00% | 100.00% | 93.24% |
| 9 | MoonshotAI: Kimi K2.5 | 96.63% | 83.17% | 100.00% | 100.00% | 100.00% | 100.00% | 91.04% |
| 10 | Qwen 3.5 35B | 96.42% | 82.08% | 100.00% | 100.00% | 100.00% | 100.00% | 88.00% |
| 11 | Z.AI GLM 5 Turbo | 96.36% | 95.82% | 100.00% | 100.00% | 96.00% | 90.00% | 94.27% |
| 12 | Qwen 3.5 122B | 96.36% | 81.81% | 100.00% | 100.00% | 100.00% | 100.00% | 91.53% |
| 13 | o4 Mini | 96.31% | 90.72% | 98.35% | 100.00% | 92.50% | 100.00% | 88.35% |
| 14 | GPT-5.2 | 96.22% | 84.32% | 99.77% | 100.00% | 97.00% | 100.00% | 90.26% |
| 15 | Gemini 3 Pro (Preview) | 96.14% | 80.75% | 99.97% | 100.00% | 100.00% | 100.00% | 88.79% |
| 16 | Qwen 3.5 Flash | 96.11% | 80.57% | 100.00% | 100.00% | 100.00% | 100.00% | 86.38% |
| 17 | Qwen 3.5 27B | 95.67% | 85.82% | 92.55% | 100.00% | 100.00% | 100.00% | 90.85% |
| 18 | MiniMax M2.7 | 95.50% | 94.05% | 95.47% | 100.00% | 88.00% | 100.00% | 89.10% |
| 19 | Grok 4.20 (Beta, Reasoning) | 95.41% | 77.05% | 100.00% | 100.00% | 100.00% | 100.00% | 91.49% |
| 20 | GPT-5.1 | 95.33% | 85.32% | 99.80% | 100.00% | 91.50% | 100.00% | 92.54% |
| 21 | GPT-5.4 (Reasoning, Low) | 95.32% | 76.62% | 100.00% | 100.00% | 100.00% | 100.00% | 91.41% |
| 22 | Nemotron 3 Super | 95.29% | 93.51% | 97.96% | 100.00% | 85.00% | 100.00% | 84.56% |
| 23 | GPT-5.4 Mini (Reasoning) | 94.44% | 81.72% | 100.00% | 100.00% | 90.50% | 100.00% | 90.65% |
| 24 | Z.AI GLM 4.7 | 94.31% | 78.57% | 99.99% | 100.00% | 93.00% | 100.00% | 88.69% |
| 25 | Z.AI GLM 5 | 94.11% | 75.55% | 100.00% | 100.00% | 95.00% | 100.00% | 91.23% |
| 26 | Qwen 3.5 9B | 94.02% | 74.59% | 100.00% | 100.00% | 95.50% | 100.00% | 86.05% |
| 27 | Gemini 3.1 Flash Lite (Preview) | 94.00% | 89.38% | 98.60% | 100.00% | 82.00% | 100.00% | 85.87% |
| 28 | GPT-5 Nano | 93.91% | 86.54% | 97.99% | 100.00% | 85.00% | 100.00% | 82.60% |
| 29 | GPT-5 | 93.53% | 96.13% | 100.00% | 100.00% | 76.50% | 95.00% | 91.93% |
| 30 | GPT-5.4 Nano (Reasoning) | 93.34% | 84.86% | 95.83% | 100.00% | 91.00% | 95.00% | 81.36% |
| 31 | Inception Mercury 2 | 92.86% | 93.76% | 97.02% | 100.00% | 73.50% | 100.00% | 83.85% |
| 32 | Stealth: Aurora Alpha | 92.59% | 88.58% | 97.37% | 100.00% | 77.00% | 100.00% | 83.79% |
| 33 | ByteDance Seed 2.0 Lite | 92.23% | 73.68% | 99.96% | 100.00% | 87.50% | 100.00% | 84.80% |
| 34 | Gemini 2.5 Pro | 92.18% | 66.47% | 99.43% | 100.00% | 100.00% | 95.00% | 88.53% |
| 35 | ByteDance Seed 2.0 Mini | 91.88% | 73.92% | 99.96% | 100.00% | 85.50% | 100.00% | 86.91% |
| 36 | GPT-5.4 Nano (Reasoning, Low) | 91.42% | 73.42% | 93.19% | 100.00% | 90.50% | 100.00% | 79.48% |
| 37 | Aion 2.0 | 90.91% | 65.48% | 89.10% | 100.00% | 100.00% | 100.00% | 89.21% |
| 38 | ByteDance Seed 1.6 | 90.83% | 74.14% | 100.00% | 100.00% | 80.00% | 100.00% | 90.70% |
| 39 | Claude Opus 4.6 | 90.72% | 83.65% | 99.93% | 100.00% | 70.00% | 100.00% | 92.35% |
| 40 | GPT-4.1 | 90.57% | 81.33% | 86.01% | 100.00% | 85.50% | 100.00% | 88.68% |
| 41 | MiniMax M2.5 | 90.42% | 86.03% | 75.58% | 100.00% | 90.50% | 100.00% | 88.71% |
| 42 | Claude Opus 4.5 | 89.84% | 80.83% | 95.88% | 100.00% | 72.50% | 100.00% | 89.69% |
| 43 | Grok 4 | 89.67% | 75.34% | 88.00% | 100.00% | 85.00% | 100.00% | 88.12% |
| 44 | Gemini 2.5 Flash Lite (Reasoning) | 89.63% | 55.13% | 96.54% | 100.00% | 96.50% | 100.00% | 85.75% |
| 45 | Z.AI GLM 4.7 Flash | 88.98% | 68.93% | 100.00% | 100.00% | 81.00% | 95.00% | 84.82% |
| 46 | Claude Opus 4 | 88.81% | 74.37% | 80.19% | 100.00% | 89.50% | 100.00% | 87.69% |
| 47 | Z.AI GLM 4.6 | 88.58% | 47.21% | 97.70% | 100.00% | 98.00% | 100.00% | 89.11% |
| 48 | Claude Sonnet 4.6 | 88.52% | 71.94% | 94.66% | 100.00% | 76.00% | 100.00% | 91.15% |
| 49 | GPT-5.4 Mini (Reasoning, Low) | 88.49% | 75.01% | 99.92% | 100.00% | 67.50% | 100.00% | 85.75% |
| 50 | Llama 3.1 Nemotron 70B | 88.31% | 47.07% | 99.99% | 100.00% | 94.50% | 100.00% | 74.70% |
| 51 | Inception Mercury | 87.38% | 85.14% | 96.78% | 100.00% | 60.00% | 95.00% | 79.50% |
| 52 | Qwen 3.5 Plus (2026-02-15) | 86.65% | 44.24% | 100.00% | 100.00% | 89.00% | 100.00% | 85.96% |
| 53 | Gemini 3 Flash (Preview) | 86.39% | 83.97% | 99.99% | 100.00% | 48.00% | 100.00% | 85.35% |
| 54 | Nemotron 3 Nano | 86.00% | 86.70% | 92.81% | 100.00% | 50.50% | 100.00% | 77.73% |
| 55 | Mistral Small 4 (Reasoning) | 85.61% | 56.00% | 96.06% | 100.00% | 76.00% | 100.00% | 82.39% |
| 56 | Mistral Large 3 | 84.91% | 52.60% | 99.95% | 100.00% | 72.00% | 100.00% | 85.43% |
| 57 | Stealth: Hunter Alpha | 84.63% | 55.55% | 87.94% | 96.67% | 83.00% | 100.00% | 87.34% |
| 58 | ByteDance Seed 1.6 Flash | 84.16% | 48.55% | 82.27% | 100.00% | 90.00% | 100.00% | 73.27% |
| 59 | Grok 4.1 Fast | 84.12% | 60.51% | 95.61% | 100.00% | 64.50% | 100.00% | 89.55% |
| 60 | Claude Sonnet 4 | 84.02% | 53.92% | 84.17% | 100.00% | 82.00% | 100.00% | 88.72% |
| 61 | DeepSeek-V2 Chat | 83.82% | 41.72% | 97.87% | 100.00% | 79.50% | 100.00% | 84.83% |
| 62 | Claude Sonnet 4.5 | 83.78% | 58.93% | 91.98% | 100.00% | 68.00% | 100.00% | 88.03% |
| 63 | Qwen3 235B A22B Instruct 2507 | 83.15% | 40.15% | 80.58% | 100.00% | 95.00% | 100.00% | 80.10% |
| 64 | GPT-4o, May 13th (temp=0) | 83.13% | 65.85% | 71.82% | 100.00% | 78.00% | 100.00% | 85.36% |
| 65 | Claude 3.5 Haiku | 82.57% | 46.03% | 76.84% | 100.00% | 90.00% | 100.00% | 83.73% |
| 66 | GPT-4o, Aug. 6th (temp=1) | 82.44% | 77.45% | 74.24% | 100.00% | 60.50% | 100.00% | 82.62% |
| 67 | Stealth: Healer Alpha | 82.30% | 61.71% | 90.46% | 93.33% | 71.00% | 95.00% | 85.93% |
| 68 | GPT-4.1 Mini | 82.30% | 66.59% | 79.41% | 100.00% | 65.50% | 100.00% | 83.20% |
| 69 | Gemini 2.5 Flash (Reasoning) | 82.25% | 46.02% | 85.21% | 100.00% | 90.00% | 90.00% | 86.51% |
| 70 | GPT-4o Mini (temp=1) | 82.16% | 78.18% | 79.62% | 100.00% | 53.00% | 100.00% | 79.08% |
| 71 | Grok 4.20 (Beta) | 82.15% | 59.16% | 99.42% | 96.67% | 60.50% | 95.00% | 83.85% |
| 72 | GPT-4o, Aug. 6th (temp=0) | 82.11% | 77.04% | 74.00% | 100.00% | 59.50% | 100.00% | 82.45% |
| 73 | GPT-5.4 | 81.95% | 64.23% | 100.00% | 100.00% | 45.50% | 100.00% | 84.32% |
| 74 | DeepSeek V3 (2024-12-26) | 81.87% | 58.30% | 88.56% | 100.00% | 62.50% | 100.00% | 83.68% |
| 75 | Qwen 3 32B | 81.66% | 31.96% | 95.82% | 100.00% | 80.50% | 100.00% | 82.21% |
| 76 | DeepSeek V3.2 | 81.58% | 45.81% | 92.09% | 100.00% | 70.00% | 100.00% | 82.25% |
| 77 | GPT-4o Mini (temp=0) | 81.43% | 82.65% | 79.49% | 100.00% | 45.00% | 100.00% | 78.29% |
| 78 | Llama 3.1 70B | 81.03% | 56.62% | 80.53% | 100.00% | 68.00% | 100.00% | 78.40% |
| 79 | GPT-4o, May 13th (temp=1) | 80.69% | 65.90% | 74.04% | 100.00% | 63.50% | 100.00% | 83.80% |
| 80 | DeepSeek V3 (2025-03-24) | 80.62% | 58.32% | 97.96% | 93.33% | 53.50% | 100.00% | 81.99% |
| 81 | Gemini 2.5 Flash Lite | 80.14% | 53.37% | 73.32% | 100.00% | 74.00% | 100.00% | 81.08% |
| 82 | Mistral Medium 3.1 | 80.13% | 64.40% | 97.23% | 100.00% | 39.00% | 100.00% | 77.83% |
| 83 | Writer: Palmyra X5 | 79.71% | 35.61% | 79.93% | 100.00% | 83.00% | 100.00% | 79.57% |
| 84 | GPT-5.4 Mini | 79.37% | 66.13% | 98.23% | 100.00% | 32.50% | 100.00% | 82.43% |
| 85 | Gemma 3 12B | 79.28% | 46.84% | 80.06% | 100.00% | 69.50% | 100.00% | 78.41% |
| 86 | Z.AI GLM 4.5 | 79.19% | 38.11% | 74.82% | 100.00% | 83.00% | 100.00% | 86.27% |
| 87 | Ministral 3 14B | 79.03% | 47.57% | 92.10% | 100.00% | 55.50% | 100.00% | 72.54% |
| 88 | GPT-5.4 Nano | 78.57% | 47.90% | 93.46% | 100.00% | 51.50% | 100.00% | 74.40% |
| 89 | Mistral Small 4 | 78.28% | 46.07% | 85.82% | 100.00% | 59.50% | 100.00% | 76.46% |
| 90 | Gemma 3 27B | 76.82% | 45.66% | 95.44% | 100.00% | 43.00% | 100.00% | 77.85% |
| 91 | Grok 4 Fast | 76.76% | 61.39% | 89.92% | 100.00% | 32.50% | 100.00% | 86.15% |
| 92 | Claude 3.5 Sonnet | 76.75% | 53.50% | 93.24% | 100.00% | 37.00% | 100.00% | 84.24% |
| 93 | DeepSeek V3.1 | 76.65% | 47.10% | 83.32% | 93.33% | 59.50% | 100.00% | 82.39% |
| 94 | Qwen 2.5 72B | 76.43% | 51.43% | 66.70% | 100.00% | 64.00% | 100.00% | 75.46% |
| 95 | Mistral Small Creative | 76.28% | 33.57% | 96.83% | 100.00% | 51.00% | 100.00% | 73.27% |
| 96 | Llama 3.1 8B | 74.82% | 56.94% | 85.64% | 90.00% | 46.50% | 95.00% | 63.37% |
| 97 | Ministral 3 8B | 74.43% | 38.01% | 74.12% | 100.00% | 60.00% | 100.00% | 71.76% |
| 98 | Mistral Small 3.2 24B | 73.17% | 42.38% | 99.97% | 100.00% | 23.50% | 100.00% | 78.60% |
| 99 | Mistral Large | 73.04% | 31.90% | 63.80% | 100.00% | 69.50% | 100.00% | 80.15% |
| 100 | Claude Haiku 4.5 | 72.48% | 51.60% | 70.28% | 70.00% | 70.50% | 100.00% | 85.14% |
| 101 | Ministral 3 3B | 72.38% | 37.04% | 74.85% | 100.00% | 50.00% | 100.00% | 67.22% |
| 102 | LFM2 24B | 69.48% | 44.98% | 63.93% | 100.00% | 38.50% | 100.00% | 58.77% |
| 103 | Mistral Large 2 | 69.19% | 21.18% | 58.78% | 100.00% | 66.00% | 100.00% | 82.41% |
| 104 | Hermes 3 405B | 69.02% | 51.42% | 79.99% | 86.67% | 37.00% | 90.00% | 82.86% |
| 105 | Claude 3 Haiku | 68.47% | 39.32% | 61.88% | 96.67% | 44.50% | 100.00% | 71.19% |
| 106 | GPT-4.1 Nano | 68.45% | 64.57% | 75.17% | 100.00% | 52.50% | 50.00% | 71.94% |
| 107 | WizardLM 2 8x22b | 67.14% | 20.73% | 71.48% | 90.00% | 53.50% | 100.00% | 71.07% |
| 108 | Claude 3.7 Sonnet | 62.54% | 66.00% | 66.35% | 33.33% | 47.00% | 100.00% | 83.39% |
| 109 | Gemini 2.5 Flash | 61.45% | 38.47% | 64.79% | 100.00% | 54.00% | 50.00% | 80.60% |
| 110 | Hermes 3 70B | 61.15% | 28.43% | 58.14% | 66.67% | 52.50% | 100.00% | 72.57% |
| 111 | Arcee AI: Trinity Large (Preview) | 60.74% | 43.69% | 57.16% | 73.33% | 44.50% | 85.00% | 73.33% |
| 112 | Gemma 3 4B | 60.30% | 22.12% | 95.04% | 53.33% | 31.00% | 100.00% | 68.57% |
| 113 | Arcee AI: Trinity Mini | 59.94% | 26.25% | 43.97% | 80.00% | 49.50% | 100.00% | 70.90% |
| 114 | Cohere Command R+ (Aug. 2024) | 59.51% | 35.13% | 60.43% | 90.00% | 42.00% | 70.00% | 69.03% |
| 115 | Mistral NeMO | 51.55% | 15.70% | 47.57% | 40.00% | 54.50% | 100.00% | 65.04% |
| 116 | Ministral 3B | 49.17% | 17.87% | 62.17% | 33.33% | 47.50% | 85.00% | 61.29% |
| 117 | Rocinante 12B | 48.47% | 14.07% | 53.12% | 36.67% | 43.50% | 95.00% | 54.55% |
| 118 | Ministral 8B | 46.82% | 21.93% | 60.35% | 33.33% | 33.50% | 85.00% | 64.87% |