LightOn Releases Fully Open Multilingual Retrieval Models mDenseOn & mLateOn

antoine_chaffin · x · 2026-07-30

The LightOn team has released mDenseOn and mLateOn, next-generation fully open models for multilingual, long-context, and code retrieval. The release includes 2.8B training pairs and 16.3M fine-tuning samples, along with the training code.

The models demonstrate that the English frontier recipe can be generalized to multilingual scenarios through translation and late interaction, proving the effectiveness of the translate-train paradigm.

Related event: LightOn Releases Open-Source Multilingual Retrieval Models with 2.8B Training Pairs(10 posts)→

Original post →

More from Models

Models channel →