Benchmark Almanac

Maritaca AI

We track 3 benchmarks that Maritaca AI made or helped make.

0 in use 1 with public code 2 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
BRoverbs Models select the Brazilian proverb that fits a story or the story that illustrates a proverb. Both directions offer five choices. Reasoning in many languages Sep 2025 95% Not in cards
OAB-Bench Models write answers for the second phase of the Brazilian bar examination. The port averages model judges’ scores against grading guidelines. Law Apr 2025 9.39 Not in cards
IMDb-PT Models classify the sentiment of Brazilian Portuguese translations of IMDb movie reviews. Reading in many languages Not in cards