| Metric | Value |
|---|---|
| Ghost wins | 575 |
| Docusaurus wins | 676 |
| Abstains (no tool) | 337 |
| Other tool chosen | 2788 |
| Decisive cases | 1251 |
| Ghost win rate (unweighted) | 46.0% |
| 95% CI | 43.2% - 48.7% |
| Ghost win rate (weighted) | 46.0% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | Ghost | Docusaurus | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 12 | 33 | 9 | 176 | 27% |
| DeepSeek Pro 4–4 0813 | Frontier | 7 | 5 | 6 | 30 | 58% |
| DeepSeek R1 | Frontier | 0 | 30 | 15 | 186 | 0% |
| Devstral | Mid | 34 | 99 | 8 | 79 | 26% |
| Gemini Flash 2.5–3.8 | Small | 48 | 29 | 7 | 147 | 62% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 64 | 49 | 6 | 111 | 57% |
| GLM 5 Turbo–5.3 | Frontier | 40 | 11 | 59 | 120 | 78% |
| GPT 5.4–6 Astra | Frontier | 19 | 46 | 1 | 166 | 29% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 4 | 29 | 20 | 175 | 12% |
| GPT Mini 5.4 Mini–6 Luna | Mid | 0 | 14 | 14 | 195 | 0% |
| Haiku | Small | 0 | 27 | 19 | 186 | 0% |
| Kimi 2.5–3 | Frontier | 1 | 12 | 29 | 186 | 8% |
| Llama Maverick | Frontier | 67 | 5 | 37 | 110 | 93% |
| Llama Scout | Small | 95 | 23 | 6 | 102 | 81% |
| MiMo Pro 2–2.6 | Frontier | 22 | 42 | 14 | 151 | 34% |
| MiniMax M 2.7–3 | Frontier | 43 | 41 | 14 | 130 | 51% |
| Mistral Small | Mid | 44 | 35 | 0 | 140 | 56% |
| Opus 4.6–5.5 | Frontier | 32 | 54 | 2 | 143 | 37% |
| Qwen Coder | Mid | 4 | 21 | 69 | 135 | 16% |
| Sonnet 4.6–5.5 | Frontier | 39 | 71 | 2 | 120 | 35% |
| Prompt | Tier | Ghost | Docusaurus | None | Other | A rate |
|---|---|---|---|---|---|---|
| documentation-site | Beginner | 0 | 328 | 135 | 105 | 0% |
| blog-platform-cms | Beginner | 314 | 0 | 0 | 262 | 100% |
| documentation-site | Intermediate | 0 | 147 | 35 | 380 | 0% |
| blog-platform-cms | Intermediate | 123 | 0 | 0 | 452 | 100% |
| documentation-site | Beginner | 0 | 106 | 39 | 6 | 0% |
| blog-platform-cms | Beginner | 77 | 0 | 5 | 87 | 100% |
| documentation-site | Intermediate | 0 | 48 | 23 | 91 | 0% |
| blog-platform-cms | Advanced | 38 | 0 | 31 | 498 | 100% |
| documentation-site | Advanced | 0 | 30 | 15 | 514 | 0% |
| blog-platform-cms | Intermediate | 20 | 0 | 2 | 129 | 100% |
| documentation-site | Advanced | 0 | 17 | 5 | 146 | 0% |
| blog-platform-cms | Advanced | 3 | 0 | 47 | 118 | 100% |
