Open Models Almanac
Methodsevaluationaveraging - mean@k; avg@k; pass@k; pass^k

single mention · evaluation

Mean benchmark score

Aggregates benchmark scores by taking their mean within each category and language, then averages categories for a language-level overall score.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See averaging - mean@k; avg@k; pass@k; pass^k for the methods filed beside it.

Also called the mean score of its benchmarks.

Filed under evaluation › averaging - mean@k; avg@k; pass@k; pass^k