| Metric | Value |
|---|---|
| Merge wins | 302 |
| Next.js wins | 358 |
| Abstains (no tool) | 896 |
| Other tool chosen | 5023 |
| Decisive cases | 660 |
| Merge win rate (unweighted) | 45.8% |
| 95% CI | 42.0% - 49.6% |
| Merge win rate (weighted) | 45.8% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Merge | Next.js | None | Other | A rate |
|---|---|---|---|---|---|---|
| Gemini 2.5 Pro | Frontier | 36 | 29 | 30 | 260 | 55% |
| GPT 5.3 Codex | Frontier | 34 | 19 | 17 | 285 | 64% |
| Claude Sonnet 4.6 | Frontier | 31 | 22 | 14 | 286 | 58% |
| Kimi K2.5 | Frontier | 46 | 3 | 35 | 156 | 94% |
| Claude Haiku 4.5 | Small | 0 | 41 | 18 | 272 | 0% |
| DeepSeek V3.2 | Mid | 0 | 37 | 26 | 183 | 0% |
| GPT 5.4 | Frontier | 19 | 17 | 22 | 188 | 53% |
| Claude Opus 4.6 | Frontier | 0 | 32 | 13 | 206 | 0% |
| MiMo V2 Pro | Frontier | 14 | 15 | 24 | 198 | 48% |
| MiniMax M2.7 | Frontier | 1 | 26 | 25 | 183 | 4% |
| GLM 5 Turbo | Frontier | 25 | 1 | 142 | 83 | 96% |
| GPT 5.5 | Frontier | 11 | 15 | 10 | 67 | 42% |
| DeepSeek R1 0528 | Frontier | 2 | 22 | 84 | 246 | 8% |
| GPT 5.4 Mini | Mid | 3 | 20 | 17 | 269 | 13% |
| Gemini 3.5 Flash | Small | 17 | 4 | 35 | 45 | 81% |
| Mistral Small 4 | Mid | 0 | 19 | 12 | 291 | 0% |
| MiniMax M3 | Frontier | 16 | 1 | 13 | 72 | 94% |
| MiMo V2.5 Pro | Frontier | 8 | 6 | 11 | 78 | 57% |
| Kimi K2.7 Code | Frontier | 11 | 1 | 10 | 82 | 92% |
| DeepSeek V4 Pro | Frontier | 9 | 2 | 32 | 57 | 82% |
| Claude Opus 4.8 | Frontier | 8 | 1 | 36 | 58 | 89% |
| GLM 5.2 | Frontier | 7 | 2 | 11 | 84 | 78% |
| Devstral 2 2512 | Mid | 2 | 6 | 49 | 261 | 25% |
| DeepSeek V4 Flash | Mid | 2 | 4 | 14 | 75 | 33% |
| Llama 4 Scout | Small | 0 | 5 | 30 | 287 | 0% |
| Llama 4 Maverick | Frontier | 0 | 4 | 17 | 323 | 0% |
| Qwen3 Coder Next | Mid | 0 | 4 | 113 | 218 | 0% |
| Gemini 2.5 Flash | Small | 0 | 0 | 36 | 210 | n/a |
| Prompt | Tier | Merge | Next.js | None | Other | A rate |
|---|---|---|---|---|---|---|
| url-shortener | Beginner | 0 | 183 | 125 | 214 | 0% |
| ai-revenue-ops-copilot | Beginner | 164 | 0 | 26 | 304 | 100% |
| ai-revenue-ops-copilot | Intermediate | 90 | 0 | 97 | 320 | 100% |
| url-shortener | Intermediate | 0 | 69 | 93 | 360 | 0% |
| ai-revenue-ops-copilot | Advanced | 47 | 0 | 90 | 371 | 100% |
| weather-dashboard | Advanced | 0 | 46 | 6 | 463 | 0% |
| multi-tenant-crm | Beginner | 0 | 25 | 148 | 352 | 0% |
| weather-dashboard | Intermediate | 0 | 14 | 1 | 518 | 0% |
| full-stack-web-app | Beginner | 0 | 8 | 74 | 53 | 0% |
| multi-tenant-crm | Intermediate | 1 | 5 | 112 | 398 | 17% |
| full-stack-web-app | Intermediate | 0 | 5 | 35 | 68 | 0% |
| full-stack-web-app | Advanced | 0 | 2 | 11 | 95 | 0% |
| multi-tenant-crm | Advanced | 0 | 1 | 35 | 493 | 0% |
| url-shortener | Advanced | 0 | 0 | 43 | 478 | n/a |
| weather-dashboard | Beginner | 0 | 0 | 0 | 536 | n/a |