Explains row-level and set-level checks
Uniqueness, overlap and references can only be judged across rows.
Can AI do this?
AI does this partly Some of the tested AI models master this; others do not show it. 2 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 44% | Not shown |
| gemini-3.1-pro-preview* | Flagship | 100% | Mastered |
| gpt-6-astra* | Flagship | 99% | Mastered |
| claude-fable-5-1* | Flagship | 100% | Mastered |
| gemini-3.8-flash* | Mid tier | 100% | Mastered |
| gpt-6.1-sol* | Mid tier | 100% | Mastered |
| claude-sonnet-5-5* | Mid tier | 99% | Mastered |
| gemini-3.5-flash-lite | Low cost | 44% | Not shown |
| gpt-6-luna | Low cost | 98% | Mastered |
| claude-haiku-5-5 | Low cost | 98% | Mastered |
* From the earlier benchmark of 2026-10: 10 data-modelling test tasks, 3 runs per task, graded by claude-opus-5-5. These models have not taken the current 123 tasks yet; the two methods differ, so compare their figures with care. They are left out of the “most models” counts and the best-model picks.
Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.
What it looks like in practice
Expected result: Uniqueness, overlap and references can only be judged across rows
How it is assessed: Classifies sample rules as row or set checks
Related knobits
Needed for
All 30 knobits of Derive data quality rules.
Source: Data-management practice used by the stage 3 reference and review checklist; formerly data-modelling K-754