Three things make a quality measurement trustworthy: a standard that is written down before the measurement, a process that repeats and reports the median rather than a single lucky run, and metrics that say what they deliberately ignore.
Truverai reports eight metrics and 32 sub-metrics, runs each evaluation three times, and publishes the configuration used so the result can be reproduced.
Truverai scores are reproducible within a published lock: engine v4.7, prompt hash 1a6c25c7e25ddbf8, model google/gemini-3-flash-preview at temperature 0, and a median of three runs. That is reproducibility within a fixed configuration, not a claim that a different language model would return the same number.
This page needs JavaScript for the full interactive view. The summary above is the same information in plain HTML.