Atlas

Benchmarks

← All benchmarks

BOUQuET (chrF++)

Chat & Writing

BOUQuET is a handcrafted, multicentric and multi-register translation benchmark, written first in nine non-English pivot languages and organized in paragraphs of varying length. Output is scored with chrF++ against human references, and this score averages the out-of-English and into-English directions over the evaluated languages.

Top models (higher is better)

ModelScore
Muse Glimmer 30B36.5
Loading Atlas data…