| Metric | Value |
|---|---|
| Courier wins | 112 |
| Amazon SNS wins | 152 |
| Abstains (no tool) | 94 |
| Other tool chosen | 6029 |
| Decisive cases | 264 |
| Courier win rate (unweighted) | 42.4% |
| 95% CI | 36.6% - 48.5% |
| Courier win rate (weighted) | 42.4% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Courier | Amazon SNS | None | Other | A rate |
|---|---|---|---|---|---|---|
| Gemini 2.5 Flash | Small | 3 | 37 | 8 | 203 | 8% |
| Claude Haiku 4.5 | Small | 0 | 36 | 0 | 287 | 0% |
| DeepSeek V3.2 | Mid | 32 | 0 | 12 | 206 | 100% |
| MiMo V2 Pro | Frontier | 0 | 31 | 4 | 217 | 0% |
| Gemini 3.5 Flash | Small | 20 | 0 | 0 | 64 | 100% |
| GPT 5.3 Codex | Frontier | 11 | 7 | 0 | 317 | 61% |
| Gemini 2.5 Pro | Frontier | 15 | 2 | 2 | 313 | 88% |
| Devstral 2 2512 | Mid | 0 | 12 | 0 | 298 | 0% |
| DeepSeek R1 0528 | Frontier | 4 | 4 | 22 | 304 | 50% |
| DeepSeek V4 Pro | Frontier | 6 | 1 | 0 | 77 | 86% |
| GPT 5.4 Mini | Mid | 2 | 5 | 8 | 318 | 29% |
| Llama 4 Scout | Small | 0 | 7 | 3 | 321 | 0% |
| Claude Opus 4.6 | Frontier | 0 | 5 | 0 | 247 | 0% |
| DeepSeek V4 Flash | Mid | 4 | 0 | 0 | 78 | 100% |
| Kimi K2.5 | Frontier | 4 | 0 | 5 | 241 | 100% |
| MiMo V2.5 Pro | Frontier | 0 | 4 | 2 | 77 | 0% |
| Claude Sonnet 4.6 | Frontier | 3 | 0 | 0 | 333 | 100% |
| GPT 5.5 | Frontier | 3 | 0 | 0 | 81 | 100% |
| Kimi K2.7 Code | Frontier | 3 | 0 | 0 | 81 | 100% |
| MiniMax M2.7 | Frontier | 2 | 1 | 12 | 231 | 67% |
| Claude Opus 4.8 | Frontier | 0 | 0 | 0 | 82 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 14 | 238 | n/a |
| GLM 5.2 | Frontier | 0 | 0 | 0 | 84 | n/a |
| GPT 5.4 | Frontier | 0 | 0 | 0 | 252 | n/a |
| Llama 4 Maverick | Frontier | 0 | 0 | 1 | 334 | n/a |
| MiniMax M3 | Frontier | 0 | 0 | 0 | 84 | n/a |
| Mistral Small 4 | Mid | 0 | 0 | 1 | 326 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 0 | 335 | n/a |
| Prompt | Tier | Courier | Amazon SNS | None | Other | A rate |
|---|---|---|---|---|---|---|
| social-media-platform | Advanced | 41 | 71 | 4 | 418 | 37% |
| chat-application | Advanced | 19 | 31 | 6 | 478 | 38% |
| project-management-tool | Advanced | 8 | 30 | 4 | 477 | 21% |
| restaurant-reservation-system | Advanced | 25 | 2 | 19 | 488 | 93% |
| chat-application | Intermediate | 0 | 11 | 3 | 520 | 0% |
| social-media-platform | Beginner | 6 | 4 | 6 | 521 | 60% |
| social-media-platform | Intermediate | 7 | 2 | 1 | 524 | 78% |
| project-management-tool | Beginner | 2 | 1 | 21 | 510 | 67% |
| restaurant-reservation-system | Intermediate | 2 | 0 | 2 | 533 | 100% |
| project-management-tool | Intermediate | 1 | 0 | 12 | 510 | 100% |
| restaurant-reservation-system | Beginner | 1 | 0 | 3 | 531 | 100% |
| chat-application | Beginner | 0 | 0 | 13 | 519 | n/a |