all the models — AI benchmark observatory
← Benchmarks

FActScore

A fine-grained atomic evaluation metric for factual precision in long-form text generation that breaks generated text into atomic facts and computes the percentage supported by reliable knowledge sources, with automated assessment using retrieval and language models

id factscore · max 1 · 3 models reported

#ModelScore
1Grok-4.1
xAI
0.97
FActScore Leaderboard · all the models