Concordance runs your question through three AI models simultaneously. A fourth synthesizes them and outputs a consensus score (0–100). High = models agree. Low = at least one is guessing. Built for the moments when one model sounds confidently wrong.
