The OPUS-MT benchmark is a systematic collection of results from these models, focusing on verifiable translation performance and large coverage in terms of languages and domains. The OPUS-MT Dashboard is a web-based platform that provides a comprehensive overview of these open translation models. It includes summaries of benchmarks for over 2,300 models covering 4,560 language directions and 294 languages. The aim is to centralize, reproduce, and cover MT evaluation combined with scalability.
Paper | Code | Results | Date | Stars |
---|