| Metric | Value |
|---|---|
| Metabase wins | 612 |
| Mixpanel wins | 1499 |
| Abstains (no tool) | 460 |
| Other tool chosen | 5396 |
| Decisive cases | 2111 |
| Metabase win rate (unweighted) | 29.0% |
| 95% CI | 27.1% - 31.0% |
| Metabase win rate (weighted) | 29.0% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Metabase | Mixpanel | None | Other | A rate |
|---|---|---|---|---|---|---|
| Llama 4 Maverick | Frontier | 0 | 282 | 0 | 137 | 0% |
| Devstral 2 2512 | Mid | 1 | 191 | 0 | 196 | 1% |
| Claude Sonnet 4.6 | Frontier | 83 | 86 | 1 | 250 | 49% |
| Gemini 2.5 Flash | Small | 0 | 161 | 21 | 130 | 0% |
| Claude Haiku 4.5 | Small | 12 | 146 | 6 | 243 | 8% |
| MiMo V2 Pro | Frontier | 55 | 81 | 24 | 155 | 40% |
| Qwen3 Coder Next | Mid | 86 | 40 | 5 | 288 | 68% |
| DeepSeek R1 0528 | Frontier | 19 | 90 | 150 | 159 | 17% |
| DeepSeek V3.2 | Mid | 3 | 104 | 26 | 179 | 3% |
| Claude Opus 4.6 | Frontier | 49 | 46 | 1 | 219 | 52% |
| Mistral Small 4 | Mid | 18 | 55 | 3 | 326 | 25% |
| GPT 5.3 Codex | Frontier | 45 | 25 | 1 | 348 | 64% |
| Llama 4 Scout | Small | 0 | 70 | 25 | 323 | 0% |
| MiniMax M2.7 | Frontier | 37 | 26 | 36 | 205 | 59% |
| GPT 5.4 Mini | Mid | 31 | 16 | 7 | 365 | 66% |
| MiMo V2.5 Pro | Frontier | 19 | 23 | 13 | 50 | 45% |
| GPT 5.4 | Frontier | 15 | 19 | 0 | 278 | 44% |
| Kimi K2.5 | Frontier | 29 | 3 | 23 | 245 | 91% |
| GLM 5 Turbo | Frontier | 27 | 3 | 63 | 222 | 90% |
| DeepSeek V4 Flash | Mid | 16 | 7 | 5 | 75 | 70% |
| Gemini 2.5 Pro | Frontier | 9 | 14 | 23 | 370 | 39% |
| Kimi K2.7 Code | Frontier | 16 | 5 | 3 | 81 | 76% |
| DeepSeek V4 Pro | Frontier | 14 | 2 | 18 | 70 | 88% |
| GPT 5.5 | Frontier | 14 | 2 | 2 | 83 | 88% |
| GLM 5.2 | Frontier | 8 | 0 | 1 | 96 | 100% |
| Claude Opus 4.8 | Frontier | 6 | 0 | 0 | 98 | 100% |
| Gemini 3.5 Flash | Small | 0 | 1 | 0 | 104 | 0% |
| MiniMax M3 | Frontier | 0 | 1 | 3 | 101 | 0% |
| Prompt | Tier | Metabase | Mixpanel | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-revenue-ops-copilot | Beginner | 261 | 64 | 5 | 190 | 80% |
| multi-tenant-crm | Intermediate | 142 | 125 | 33 | 231 | 53% |
| fitness-tracking-app | Intermediate | 1 | 218 | 17 | 301 | 0% |
| multi-tenant-crm | Beginner | 23 | 148 | 47 | 315 | 13% |
| ai-revenue-ops-copilot | Advanced | 81 | 83 | 15 | 329 | 49% |
| ai-revenue-ops-copilot | Intermediate | 8 | 156 | 2 | 363 | 5% |
| saas-application | Beginner | 0 | 137 | 18 | 382 | 0% |
| fitness-tracking-app | Beginner | 0 | 135 | 1 | 400 | 0% |
| multi-tenant-crm | Advanced | 90 | 26 | 37 | 383 | 78% |
| saas-application | Intermediate | 0 | 104 | 51 | 380 | 0% |
| fitness-tracking-app | Advanced | 5 | 78 | 13 | 429 | 6% |
| saas-application | Advanced | 1 | 79 | 31 | 425 | 1% |
| url-shortener | Intermediate | 0 | 61 | 75 | 400 | 0% |
| url-shortener | Beginner | 0 | 44 | 110 | 381 | 0% |
| url-shortener | Advanced | 0 | 41 | 5 | 487 | 0% |