Benchmark

WMT24++

Measures: English-to-X machine translation into 55 languages and dialects (news, social, speech, literary domains) with human post-edited references; scored with automatic metrics such as XCOMET-XXL or MetricX.

Maintainer
Google (Deutsch et al.), extending the WMT24 shared task
Introduced
2025
Status
Active
Saturation
Near-ceiling
Typical provenance
Self-reported
In scoring set
No

Paper

Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.

No results to show.

WMT24++: leaderboard and what it measures · supermarket.dev