| Metric | Value |
|---|---|
| Merge wins | 206 |
| tRPC wins | 306 |
| Abstains (no tool) | 881 |
| Other tool chosen | 4581 |
| Decisive cases | 512 |
| Merge win rate (unweighted) | 40.2% |
| 95% CI | 36.1% - 44.5% |
| Merge win rate (weighted) | 40.2% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Merge | tRPC | None | Other | A rate |
|---|---|---|---|---|---|---|
| Claude Sonnet 4.6 | Frontier | 23 | 70 | 12 | 218 | 25% |
| Kimi K2.5 | Frontier | 48 | 40 | 51 | 153 | 55% |
| Claude Opus 4.6 | Frontier | 0 | 56 | 17 | 235 | 0% |
| Gemini 2.5 Pro | Frontier | 30 | 20 | 33 | 239 | 60% |
| GPT 5.4 Mini | Mid | 2 | 43 | 21 | 224 | 4% |
| GLM 5 Turbo | Frontier | 27 | 9 | 168 | 104 | 75% |
| GPT 5.4 | Frontier | 21 | 9 | 38 | 235 | 70% |
| GPT 5.3 Codex | Frontier | 28 | 1 | 22 | 271 | 97% |
| MiniMax M2.7 | Frontier | 1 | 27 | 40 | 222 | 4% |
| MiMo V2 Pro | Frontier | 14 | 0 | 38 | 256 | 100% |
| DeepSeek V3.2 | Mid | 0 | 12 | 36 | 254 | 0% |
| Kimi K2.7 Code | Frontier | 1 | 4 | 2 | 8 | 20% |
| Claude Haiku 4.5 | Small | 0 | 5 | 41 | 260 | 0% |
| Gemini 3.5 Flash | Small | 3 | 1 | 5 | 6 | 75% |
| MiniMax M3 | Frontier | 2 | 2 | 3 | 8 | 50% |
| GLM 5.2 | Frontier | 1 | 3 | 2 | 9 | 25% |
| GPT 5.5 | Frontier | 1 | 2 | 1 | 11 | 33% |
| Devstral 2 2512 | Mid | 2 | 0 | 55 | 257 | 100% |
| DeepSeek R1 0528 | Frontier | 1 | 0 | 80 | 242 | 100% |
| MiMo V2.5 Pro | Frontier | 1 | 0 | 2 | 12 | 100% |
| DeepSeek V4 Flash | Mid | 0 | 1 | 3 | 10 | 0% |
| Llama 4 Scout | Small | 0 | 1 | 23 | 272 | 0% |
| Claude Opus 4.8 | Frontier | 0 | 0 | 6 | 9 | n/a |
| DeepSeek V4 Pro | Frontier | 0 | 0 | 6 | 8 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 52 | 250 | n/a |
| Llama 4 Maverick | Frontier | 0 | 0 | 16 | 306 | n/a |
| Mistral Small 4 | Mid | 0 | 0 | 7 | 293 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 101 | 209 | n/a |
| Prompt | Tier | Merge | tRPC | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-revenue-ops-copilot | Beginner | 112 | 0 | 24 | 261 | 100% |
| multi-tenant-crm | Beginner | 0 | 100 | 118 | 208 | 0% |
| multi-tenant-crm | Intermediate | 1 | 73 | 89 | 256 | 1% |
| ai-revenue-ops-copilot | Intermediate | 60 | 0 | 82 | 268 | 100% |
| url-shortener | Intermediate | 0 | 34 | 74 | 316 | 0% |
| ai-revenue-ops-copilot | Advanced | 33 | 0 | 75 | 303 | 100% |
| url-shortener | Beginner | 0 | 28 | 99 | 302 | 0% |
| multi-tenant-crm | Beginner | 0 | 18 | 60 | 16 | 0% |
| multi-tenant-crm | Intermediate | 0 | 18 | 40 | 36 | 0% |
| weather-dashboard | Advanced | 0 | 10 | 2 | 82 | 0% |
| full-stack-web-app | Intermediate | 0 | 6 | 5 | 6 | 0% |
| url-shortener | Intermediate | 0 | 6 | 33 | 55 | 0% |
| url-shortener | Advanced | 0 | 3 | 15 | 75 | 0% |
| weather-dashboard | Advanced | 0 | 3 | 4 | 414 | 0% |
| full-stack-web-app | Advanced | 0 | 2 | 0 | 14 | 0% |
| multi-tenant-crm | Advanced | 0 | 2 | 32 | 399 | 0% |
| url-shortener | Advanced | 0 | 2 | 38 | 390 | 0% |
| full-stack-web-app | Beginner | 0 | 1 | 13 | 5 | 0% |
| multi-tenant-crm | Advanced | 0 | 0 | 9 | 86 | n/a |
| url-shortener | Beginner | 0 | 0 | 66 | 29 | n/a |
| weather-dashboard | Beginner | 0 | 0 | 0 | 95 | n/a |
| weather-dashboard | Intermediate | 0 | 0 | 2 | 93 | n/a |
| weather-dashboard | Beginner | 0 | 0 | 0 | 438 | n/a |
| weather-dashboard | Intermediate | 0 | 0 | 1 | 434 | n/a |