| Metric | Value |
|---|---|
| Pusher wins | 1008 |
| Firebase wins | 1186 |
| Abstains (no tool) | 75 |
| Other tool chosen | 4309 |
| Decisive cases | 2194 |
| Pusher win rate (unweighted) | 45.9% |
| 95% CI | 43.9% - 48.0% |
| Pusher win rate (weighted) | 45.9% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | Pusher | Firebase | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 28 | 8 | 2 | 306 | 78% |
| DeepSeek Pro 4–4 0813 | Frontier | 2 | 7 | 1 | 60 | 22% |
| DeepSeek R1 | Frontier | 65 | 64 | 50 | 165 | 50% |
| Devstral | Mid | 12 | 294 | 0 | 22 | 4% |
| Gemini Flash 2.5–3.8 | Small | 56 | 92 | 3 | 195 | 38% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 146 | 0 | 0 | 199 | 100% |
| GLM 5 Turbo–5.3 | Frontier | 145 | 0 | 4 | 197 | 100% |
| GPT 5.4–6 Astra | Frontier | 27 | 10 | 0 | 305 | 73% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 11 | 2 | 0 | 332 | 85% |
| GPT Mini 5.4 Mini–6 Luna | Mid | 16 | 27 | 6 | 293 | 37% |
| Haiku | Small | 2 | 89 | 0 | 248 | 2% |
| Kimi 2.5–3 | Frontier | 15 | 0 | 5 | 323 | 100% |
| Llama Maverick | Frontier | 30 | 0 | 0 | 312 | 100% |
| Llama Scout | Small | 52 | 99 | 1 | 187 | 34% |
| MiMo Pro 2–2.6 | Frontier | 129 | 95 | 1 | 119 | 58% |
| MiniMax M 2.7–3 | Frontier | 33 | 58 | 1 | 251 | 36% |
| Mistral Small | Mid | 47 | 145 | 0 | 149 | 24% |
| Opus 4.6–5.5 | Frontier | 46 | 0 | 1 | 298 | 100% |
| Qwen Coder | Mid | 131 | 196 | 0 | 17 | 40% |
| Sonnet 4.6–5.5 | Frontier | 15 | 0 | 0 | 331 | 100% |
| Prompt | Tier | Pusher | Firebase | None | Other | A rate |
|---|---|---|---|---|---|---|
| social-media-platform | Intermediate | 175 | 101 | 6 | 288 | 63% |
| chat-application | Intermediate | 39 | 199 | 9 | 327 | 16% |
| project-management-tool | Beginner | 82 | 150 | 7 | 335 | 35% |
| social-media-platform | Beginner | 129 | 99 | 23 | 315 | 57% |
| chat-application | Beginner | 22 | 166 | 4 | 380 | 12% |
| project-management-tool | Intermediate | 59 | 126 | 4 | 384 | 32% |
| social-media-platform | Advanced | 132 | 31 | 4 | 405 | 81% |
| chat-application | Advanced | 53 | 55 | 1 | 461 | 49% |
| social-media-platform | Beginner | 54 | 32 | 4 | 80 | 63% |
| project-management-tool | Advanced | 40 | 44 | 0 | 489 | 48% |
| chat-application | Beginner | 18 | 55 | 3 | 91 | 25% |
| social-media-platform | Intermediate | 47 | 23 | 1 | 81 | 67% |
| project-management-tool | Beginner | 38 | 25 | 4 | 85 | 60% |
| chat-application | Intermediate | 18 | 45 | 2 | 105 | 29% |
| project-management-tool | Intermediate | 26 | 25 | 0 | 101 | 51% |
| social-media-platform | Advanced | 47 | 1 | 2 | 119 | 98% |
| chat-application | Advanced | 19 | 2 | 1 | 130 | 90% |
| project-management-tool | Advanced | 10 | 7 | 0 | 133 | 59% |
