Compare the complete matrix
Start with all 17 configurations, then filter any combination of model family, reasoning tier, cost evidence, Pareto status, or exact configuration.
Selected run context: The live public read could not be read. No synthetic measurements were substituted.
Cannot read public_runs: invalid response shapeComparison workspace: The live public read could not be read. No synthetic measurements were substituted.
The exact 17-configuration score, run, and efficiency join is unavailable. No partial comparison is shown.Evidence notesHow to interpret time, cost, and uncertainty
Comparison matrix: Values come from RLS-protected public reads.
Official rows use calibrated ability and its conditional interval. Synthetic rows use descriptive quality and task-mix sensitivity and are not Official. Strict pass, adapter elapsed, and estimated Standard API-equivalent token cost remain separate, coverage-qualified dimensions; AIQ does not combine them into one rank. Summed cell time can overlap. Missing values remain unavailable, never zero. Inspect calibration evidence.