Benchmark

WMT23

The Eighth Conference on Machine Translation (WMT23) benchmark evaluating machine translation systems across 8 language pairs (14 translation directions)…

Modality
text
Categories
language, healthcare
Openness
unknown
Source
llm_stats
Reported scores
4

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Gemini 1.0 ProGoogle0.7172024-02-15evidence
Gemini 1.5 FlashGoogle0.7412024-05-01evidence
Gemini 1.5 Flash 8BGoogle0.7262024-03-15evidence
Gemini 1.5 ProGoogle0.7512024-05-01evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.