Omnilingual MT: Machine Translation for 1,600 Languages

2026-03-17

AI Summary

This work addresses the significant gap between the limited language coverage of current machine translation systems—spanning only around 200 languages—and the needs of over 7,000 spoken languages worldwide, compounded by the absence of reliable evaluation benchmarks. The authors present a high-quality translation system supporting more than 1,600 languages, integrating large-scale public corpora with a newly curated, manually vetted bilingual dataset called MeDLEY. They propose two specialized large language model architectures: a decoder-only variant (OMT-LLaMA) and an encoder-decoder variant (OMT-NLLB). These models, operating at 1B–8B parameters, outperform 70B-parameter general-purpose LLMs, markedly improving generation coherence and cross-lingual understanding for low-resource languages. Additionally, the study introduces BOUQuET and Met-BOUQuET, dynamic and evolving evaluation benchmarks that establish a new paradigm for multilingual translation research.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Mechanism Design

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

High-quality machine translation (MT) can scale to hundreds of languages, setting a high bar for multilingual systems. However, compared to the world's 7,000 languages, current systems still offer only limited coverage: about 200 languages on the target side, and maybe a few hundreds more on the source side, supported due to cross-lingual transfer. And even these numbers have been hard to evaluate due to the lack of reliable benchmarks and metrics. We present Omnilingual Machine Translation (OMT), the first MT system supporting more than 1,600 languages. This scale is enabled by a comprehensive data strategy that integrates large public multilingual corpora with newly created datasets, including manually curated MeDLEY bitext. We explore two ways of specializing a Large Language model (LLM) for machine translation: as a decoder-only model (OMT-LLaMA) or as a module in an encoder-decoder architecture (OMT-NLLB). Notably, all our 1B to 8B parameter models match or exceed the MT performance of a 70B LLM baseline, revealing a clear specialization advantage and enabling strong translation quality in low-compute settings. Moreover, our evaluation of English-to-1,600 translations further shows that while baseline models can interpret undersupported languages, they frequently fail to generate them with meaningful fidelity; OMT-LLaMA models substantially expand the set of languages for which coherent generation is feasible. Additionally, OMT models improve in cross-lingual transfer, being close to solving the "understanding" part of the puzzle in MT for the 1,600 evaluated. Our leaderboard and main human-created evaluation datasets (BOUQuET and Met-BOUQuET) are dynamically evolving towards Omnilinguality and freely available.

Problem

• machine translation
• low-resource languages
• language coverage
• multilingual systems
• evaluation benchmarks

Innovation

• Omnilingual Machine Translation
• low-resource languages
• model specialization
• cross-lingual transfer
• multilingual NMT