Leading: OpenWeatherMap (70.2%)
| Metric | Value |
|---|---|
| OpenWeatherMap wins | 825 |
| FastAPI wins | 350 |
| Abstains (no tool) | 569 |
| Other tool chosen | 2872 |
| Decisive cases | 1175 |
| OpenWeatherMap win rate (unweighted) | 70.2% |
| 95% CI | 67.5% - 72.8% |
| OpenWeatherMap win rate (weighted) | 70.2% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | OpenWeatherMap | FastAPI | None | Other | A rate |
|---|---|---|---|---|---|---|
| Mistral Small 4 | Mid | 63 | 29 | 6 | 131 | 68% |
| Qwen3 Coder Next | Mid | 57 | 29 | 79 | 77 | 66% |
| GPT 5.4 | Frontier | 31 | 49 | 21 | 146 | 39% |
| DeepSeek R1 0528 | Frontier | 34 | 45 | 54 | 118 | 43% |
| MiMo V2 Pro | Frontier | 61 | 11 | 26 | 153 | 85% |
| Claude Opus 4.6 | Frontier | 50 | 20 | 11 | 171 | 71% |
| Gemini 2.5 Flash | Small | 63 | 0 | 35 | 149 | 100% |
| Llama 4 Maverick | Frontier | 63 | 0 | 4 | 184 | 100% |
| Gemini 2.5 Pro | Frontier | 33 | 29 | 16 | 172 | 53% |
| Devstral 2 2512 | Mid | 42 | 19 | 37 | 145 | 69% |
| MiniMax M2.7 | Frontier | 32 | 29 | 27 | 146 | 52% |
| GLM 5 Turbo | Frontier | 53 | 6 | 141 | 51 | 90% |
| GPT 5.3 Codex | Frontier | 30 | 28 | 9 | 183 | 52% |
| Llama 4 Scout | Small | 52 | 1 | 11 | 163 | 98% |
| DeepSeek V3.2 | Mid | 27 | 25 | 27 | 166 | 52% |
| Claude Haiku 4.5 | Small | 28 | 23 | 15 | 170 | 55% |
| GPT 5.4 Mini | Mid | 39 | 3 | 12 | 169 | 93% |
| Claude Sonnet 4.6 | Frontier | 39 | 1 | 5 | 206 | 98% |
| Kimi K2.5 | Frontier | 28 | 3 | 33 | 172 | 90% |
| Prompt | Tier | OpenWeatherMap | FastAPI | None | Other | A rate |
|---|---|---|---|---|---|---|
| weather-dashboard | Intermediate | 327 | 0 | 1 | 69 | 100% |
| weather-dashboard | Beginner | 318 | 0 | 0 | 81 | 100% |
| weather-dashboard | Advanced | 180 | 12 | 4 | 188 | 94% |
| url-shortener | Advanced | 0 | 90 | 36 | 265 | 0% |
| ai-revenue-ops-copilot | Advanced | 0 | 88 | 68 | 218 | 0% |
| url-shortener | Intermediate | 0 | 50 | 65 | 271 | 0% |
| ai-revenue-ops-copilot | Intermediate | 0 | 42 | 72 | 258 | 0% |
| ai-revenue-ops-copilot | Beginner | 0 | 40 | 24 | 297 | 0% |
| multi-tenant-crm | Intermediate | 0 | 11 | 74 | 296 | 0% |
| url-shortener | Beginner | 0 | 11 | 88 | 291 | 0% |
| multi-tenant-crm | Advanced | 0 | 5 | 30 | 359 | 0% |
| multi-tenant-crm | Beginner | 0 | 1 | 107 | 279 | 0% |