| Metric | Value |
|---|---|
| Cube.js wins | 52 |
| Hex wins | 72 |
| Abstains (no tool) | 459 |
| Other tool chosen | 7382 |
| Decisive cases | 124 |
| Cube.js win rate (unweighted) | 41.9% |
| 95% CI | 33.6% - 50.7% |
| Cube.js win rate (weighted) | 41.9% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Cube.js | Hex | None | Other | A rate |
|---|---|---|---|---|---|---|
| Mistral Small 4 | Mid | 13 | 37 | 3 | 350 | 26% |
| Gemini 2.5 Pro | Frontier | 25 | 0 | 26 | 365 | 100% |
| MiniMax M2.7 | Frontier | 8 | 4 | 36 | 271 | 67% |
| Kimi K2.5 | Frontier | 2 | 8 | 23 | 278 | 20% |
| Devstral 2 2512 | Mid | 0 | 9 | 0 | 394 | 0% |
| GPT 5.4 | Frontier | 0 | 6 | 0 | 321 | 0% |
| Claude Opus 4.6 | Frontier | 0 | 4 | 1 | 325 | 0% |
| DeepSeek V4 Pro | Frontier | 2 | 1 | 15 | 71 | 67% |
| GLM 5.2 | Frontier | 2 | 0 | 1 | 87 | 100% |
| Claude Opus 4.8 | Frontier | 0 | 1 | 0 | 88 | 0% |
| Kimi K2.7 Code | Frontier | 0 | 1 | 3 | 86 | 0% |
| Llama 4 Maverick | Frontier | 0 | 1 | 0 | 418 | 0% |
| Claude Haiku 4.5 | Small | 0 | 0 | 5 | 401 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 1 | 419 | n/a |
| DeepSeek R1 0528 | Frontier | 0 | 0 | 149 | 269 | n/a |
| DeepSeek V3.2 | Mid | 0 | 0 | 27 | 300 | n/a |
| DeepSeek V4 Flash | Mid | 0 | 0 | 4 | 84 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 22 | 305 | n/a |
| Gemini 3.5 Flash | Small | 0 | 0 | 0 | 90 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 65 | 265 | n/a |
| GPT 5.3 Codex | Frontier | 0 | 0 | 1 | 418 | n/a |
| GPT 5.4 Mini | Mid | 0 | 0 | 7 | 412 | n/a |
| GPT 5.5 | Frontier | 0 | 0 | 1 | 86 | n/a |
| Llama 4 Scout | Small | 0 | 0 | 25 | 393 | n/a |
| MiMo V2 Pro | Frontier | 0 | 0 | 25 | 305 | n/a |
| MiMo V2.5 Pro | Frontier | 0 | 0 | 11 | 79 | n/a |
| MiniMax M3 | Frontier | 0 | 0 | 3 | 87 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 5 | 415 | n/a |
| Prompt | Tier | Cube.js | Hex | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-revenue-ops-copilot | Advanced | 0 | 41 | 16 | 449 | 0% |
| multi-tenant-crm | Advanced | 35 | 0 | 37 | 464 | 100% |
| ai-revenue-ops-copilot | Beginner | 4 | 21 | 5 | 490 | 16% |
| multi-tenant-crm | Intermediate | 11 | 0 | 34 | 485 | 100% |
| ai-revenue-ops-copilot | Intermediate | 0 | 10 | 3 | 516 | 0% |
| fitness-tracking-app | Intermediate | 1 | 0 | 16 | 520 | 100% |
| saas-application | Intermediate | 1 | 0 | 52 | 482 | 100% |
| fitness-tracking-app | Beginner | 0 | 0 | 1 | 535 | n/a |
| fitness-tracking-app | Advanced | 0 | 0 | 13 | 513 | n/a |
| multi-tenant-crm | Beginner | 0 | 0 | 47 | 487 | n/a |
| saas-application | Beginner | 0 | 0 | 16 | 521 | n/a |
| saas-application | Advanced | 0 | 0 | 30 | 506 | n/a |
| url-shortener | Beginner | 0 | 0 | 109 | 427 | n/a |
| url-shortener | Advanced | 0 | 0 | 5 | 527 | n/a |
| url-shortener | Intermediate | 0 | 0 | 75 | 460 | n/a |