| Metric | Value |
|---|---|
| Mixpanel wins | 1277 |
| PostHog wins | 1755 |
| Abstains (no tool) | 379 |
| Other tool chosen | 2784 |
| Decisive cases | 3032 |
| Mixpanel win rate (unweighted) | 42.1% |
| 95% CI | 40.4% - 43.9% |
| Mixpanel win rate (weighted) | 42.1% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Mixpanel | PostHog | None | Other | A rate |
|---|---|---|---|---|---|---|
| Llama 4 Maverick | Frontier | 225 | 2 | 0 | 102 | 99% |
| GPT 5.4 Mini | Mid | 13 | 194 | 5 | 118 | 6% |
| DeepSeek V3.2 | Mid | 109 | 88 | 27 | 103 | 55% |
| GPT 5.4 | Frontier | 19 | 178 | 0 | 130 | 10% |
| GLM 5 Turbo | Frontier | 4 | 180 | 65 | 81 | 2% |
| Devstral 2 2512 | Mid | 172 | 6 | 0 | 150 | 97% |
| Gemini 2.5 Flash | Small | 167 | 7 | 22 | 131 | 96% |
| Claude Haiku 4.5 | Small | 108 | 63 | 4 | 145 | 63% |
| Claude Sonnet 4.6 | Frontier | 68 | 99 | 1 | 162 | 41% |
| Gemini 2.5 Pro | Frontier | 10 | 157 | 24 | 136 | 6% |
| GPT 5.3 Codex | Frontier | 15 | 149 | 1 | 164 | 9% |
| Qwen3 Coder Next | Mid | 28 | 135 | 5 | 162 | 17% |
| Claude Opus 4.6 | Frontier | 49 | 110 | 1 | 170 | 31% |
| Mistral Small 4 | Mid | 46 | 109 | 2 | 157 | 30% |
| Kimi K2.5 | Frontier | 3 | 152 | 23 | 133 | 2% |
| MiniMax M2.7 | Frontier | 29 | 105 | 36 | 149 | 22% |
| MiMo V2 Pro | Frontier | 84 | 11 | 25 | 210 | 88% |
| DeepSeek R1 0528 | Frontier | 77 | 10 | 117 | 124 | 89% |
| Llama 4 Scout | Small | 51 | 0 | 21 | 257 | 100% |
| Prompt | Tier | Mixpanel | PostHog | None | Other | A rate |
|---|---|---|---|---|---|---|
| saas-application | Beginner | 117 | 275 | 14 | 12 | 30% |
| saas-application | Intermediate | 84 | 265 | 47 | 20 | 24% |
| fitness-tracking-app | Intermediate | 189 | 145 | 13 | 71 | 57% |
| ai-revenue-ops-copilot | Intermediate | 131 | 200 | 3 | 77 | 40% |
| multi-tenant-crm | Beginner | 126 | 177 | 36 | 76 | 42% |
| multi-tenant-crm | Intermediate | 111 | 138 | 28 | 134 | 45% |
| saas-application | Advanced | 63 | 140 | 26 | 188 | 31% |
| fitness-tracking-app | Beginner | 111 | 87 | 1 | 218 | 56% |
| url-shortener | Intermediate | 54 | 131 | 59 | 172 | 29% |
| url-shortener | Beginner | 40 | 83 | 87 | 207 | 33% |
| ai-revenue-ops-copilot | Advanced | 74 | 33 | 14 | 274 | 69% |
| ai-revenue-ops-copilot | Beginner | 55 | 21 | 5 | 322 | 72% |
| fitness-tracking-app | Advanced | 65 | 7 | 10 | 328 | 90% |
| url-shortener | Advanced | 36 | 24 | 4 | 350 | 60% |
| multi-tenant-crm | Advanced | 21 | 29 | 32 | 335 | 42% |