| Metric | Value |
|---|---|
| Sanity wins | 649 |
| Docusaurus wins | 486 |
| Abstains (no tool) | 234 |
| Other tool chosen | 1781 |
| Decisive cases | 1135 |
| Sanity win rate (unweighted) | 57.2% |
| 95% CI | 54.3% - 60.0% |
| Sanity win rate (weighted) | 57.2% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Sanity | Docusaurus | None | Other | A rate |
|---|---|---|---|---|---|---|
| Kimi K2.5 | Frontier | 104 | 8 | 17 | 31 | 93% |
| GPT 5.4 Mini | Mid | 82 | 11 | 8 | 61 | 88% |
| GPT 5.3 Codex | Frontier | 69 | 23 | 12 | 61 | 75% |
| GPT 5.4 | Frontier | 55 | 33 | 0 | 74 | 63% |
| Claude Sonnet 4.6 | Frontier | 26 | 54 | 0 | 88 | 33% |
| Devstral 2 2512 | Mid | 0 | 78 | 5 | 85 | 0% |
| Claude Opus 4.6 | Frontier | 44 | 32 | 0 | 86 | 58% |
| Qwen3 Coder Next | Mid | 56 | 15 | 53 | 44 | 79% |
| DeepSeek V3.2 | Mid | 44 | 21 | 7 | 89 | 68% |
| Mistral Small 4 | Mid | 37 | 19 | 0 | 103 | 66% |
| Gemini 2.5 Pro | Frontier | 21 | 34 | 6 | 106 | 38% |
| MiniMax M2.7 | Frontier | 20 | 32 | 10 | 97 | 38% |
| GLM 5 Turbo | Frontier | 41 | 2 | 51 | 66 | 95% |
| MiMo V2 Pro | Frontier | 11 | 27 | 10 | 114 | 29% |
| DeepSeek R1 0528 | Frontier | 15 | 22 | 12 | 118 | 41% |
| Gemini 2.5 Flash | Small | 0 | 25 | 0 | 136 | 0% |
| Claude Haiku 4.5 | Small | 0 | 21 | 8 | 139 | 0% |
| Llama 4 Scout | Small | 0 | 16 | 4 | 143 | 0% |
| MiniMax M3 | Frontier | 4 | 1 | 0 | 1 | 80% |
| Claude Opus 4.8 | Frontier | 3 | 2 | 0 | 1 | 60% |
| Gemini 3.5 Flash | Small | 4 | 0 | 1 | 1 | 100% |
| Kimi K2.7 Code | Frontier | 4 | 0 | 1 | 0 | 100% |
| GLM 5.2 | Frontier | 3 | 1 | 0 | 2 | 75% |
| GPT 5.5 | Frontier | 2 | 2 | 0 | 2 | 50% |
| DeepSeek V4 Flash | Mid | 1 | 2 | 0 | 3 | 33% |
| MiMo V2.5 Pro | Frontier | 1 | 2 | 0 | 2 | 33% |
| Llama 4 Maverick | Frontier | 0 | 3 | 28 | 125 | 0% |
| DeepSeek V4 Pro | Frontier | 2 | 0 | 1 | 3 | 100% |
| Prompt | Tier | Sanity | Docusaurus | None | Other | A rate |
|---|---|---|---|---|---|---|
| documentation-site | Beginner | 9 | 250 | 96 | 76 | 3% |
| documentation-site | Intermediate | 77 | 110 | 26 | 214 | 41% |
| blog-platform-cms | Intermediate | 182 | 0 | 0 | 255 | 100% |
| documentation-site | Advanced | 129 | 25 | 14 | 255 | 84% |
| blog-platform-cms | Beginner | 96 | 0 | 0 | 342 | 100% |
| blog-platform-cms | Advanced | 88 | 0 | 29 | 313 | 100% |
| documentation-site | Beginner | 0 | 66 | 24 | 4 | 0% |
| documentation-site | Intermediate | 12 | 27 | 11 | 41 | 31% |
| documentation-site | Advanced | 26 | 8 | 2 | 59 | 76% |
| blog-platform-cms | Beginner | 20 | 0 | 3 | 72 | 100% |
| blog-platform-cms | Advanced | 5 | 0 | 28 | 62 | 100% |
| blog-platform-cms | Intermediate | 5 | 0 | 1 | 88 | 100% |