Evidence laboratory
Numbers need a cohort before they need a rank.
Local static measurements are route checks, not product superiority claims. A paired benchmark is required for ranking or token economy.
| Metric | Requirement for claim | Current state |
|---|---|---|
| Ranking | Same task, model, hardware, n and quality threshold | N/A |
| Token economy | 1 - tokens_with/tokens_baseline, with cohort and provider price | N/A |
| Route smoke | Versioned static build and local HTTP check | Measured locally |