Cites a catalogue metric for every panel
Each panel's metric id from the stage 2 catalogue, none invented
Can AI do this?
AI does this well Most of the tested AI models master this. 3 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 90% | Mastered |
| gemini-3.5-flash-lite | Low cost | 90% | Mastered |
| gpt-6-luna | Low cost | 90% | Mastered |
| claude-haiku-5-5 | Low cost | 90% | Not shown |
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: Each panel's metric id from the stage 2 catalogue, none invented
How it is assessed: Every panel traces to one catalogue metric
Related knobits
Builds on
All 14 knobits of Specify reports and dashboards.
Source: Graded on its own by dwbi-report-specification