| Metric | Value |
|---|---|
| Stream wins | 288 |
| Pusher wins | 680 |
| Abstains (no tool) | 52 |
| Other tool chosen | 3776 |
| Decisive cases | 968 |
| Stream win rate (unweighted) | 29.8% |
| 95% CI | 27.0% - 32.7% |
| Stream win rate (weighted) | 29.8% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Stream | Pusher | None | Other | A rate |
|---|---|---|---|---|---|---|
| Gemini 2.5 Pro | Frontier | 6 | 107 | 0 | 139 | 5% |
| Qwen3 Coder Next | Mid | 0 | 97 | 0 | 153 | 0% |
| GLM 5 Turbo | Frontier | 23 | 72 | 3 | 91 | 24% |
| MiMo V2 Pro | Frontier | 4 | 78 | 0 | 106 | 5% |
| Kimi K2.5 | Frontier | 50 | 2 | 3 | 132 | 96% |
| Claude Opus 4.6 | Frontier | 21 | 31 | 0 | 137 | 40% |
| GPT 5.4 | Frontier | 33 | 14 | 0 | 140 | 70% |
| DeepSeek R1 0528 | Frontier | 2 | 44 | 40 | 165 | 4% |
| Llama 4 Scout | Small | 0 | 40 | 0 | 208 | 0% |
| Llama 4 Maverick | Frontier | 12 | 26 | 0 | 211 | 32% |
| Gemini 2.5 Flash | Small | 30 | 7 | 0 | 152 | 81% |
| Claude Sonnet 4.6 | Frontier | 29 | 6 | 0 | 217 | 83% |
| Gemini 3.5 Flash | Small | 7 | 24 | 0 | 32 | 23% |
| Mistral Small 4 | Mid | 0 | 31 | 0 | 216 | 0% |
| GPT 5.4 Mini | Mid | 18 | 7 | 4 | 219 | 72% |
| GLM 5.2 | Frontier | 6 | 12 | 0 | 45 | 33% |
| MiMo V2.5 Pro | Frontier | 3 | 15 | 0 | 45 | 17% |
| MiniMax M2.7 | Frontier | 7 | 10 | 0 | 170 | 41% |
| MiniMax M3 | Frontier | 0 | 17 | 0 | 46 | 0% |
| Kimi K2.7 Code | Frontier | 13 | 2 | 0 | 48 | 87% |
| DeepSeek V3.2 | Mid | 6 | 8 | 1 | 174 | 43% |
| DeepSeek V4 Flash | Mid | 2 | 12 | 0 | 47 | 14% |
| Claude Opus 4.8 | Frontier | 4 | 7 | 0 | 52 | 36% |
| GPT 5.3 Codex | Frontier | 4 | 6 | 0 | 242 | 40% |
| GPT 5.5 | Frontier | 6 | 0 | 0 | 57 | 100% |
| DeepSeek V4 Pro | Frontier | 2 | 2 | 1 | 56 | 50% |
| Devstral 2 2512 | Mid | 0 | 2 | 0 | 232 | 0% |
| Claude Haiku 4.5 | Small | 0 | 1 | 0 | 244 | 0% |
| Prompt | Tier | Stream | Pusher | None | Other | A rate |
|---|---|---|---|---|---|---|
| chat-application | Beginner | 215 | 19 | 4 | 296 | 92% |
| social-media-platform | Intermediate | 7 | 163 | 4 | 357 | 4% |
| social-media-platform | Advanced | 9 | 125 | 3 | 397 | 7% |
| social-media-platform | Beginner | 11 | 120 | 22 | 374 | 8% |
| project-management-tool | Beginner | 0 | 75 | 7 | 453 | 0% |
| chat-application | Advanced | 21 | 48 | 1 | 461 | 30% |
| chat-application | Intermediate | 24 | 35 | 8 | 469 | 41% |
| project-management-tool | Intermediate | 0 | 57 | 3 | 474 | 0% |
| project-management-tool | Advanced | 1 | 38 | 0 | 495 | 3% |