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.
More from Models
- Report: Anthropic Opus 5 Incoming, Kimi K3 to Be Largest OSS Model — WolframRvnwlf · 2026-07-30
- Google Launches Gemini Robotics ER 2 with Enhanced Multi-Robot Collaboration — OfficialLoganK · 2026-07-30
- Google Launches Gemini Robotics ER 2 for Embodied Reasoning — OfficialLoganK · 2026-07-30
- Anthropic Lacks Standard Completions Endpoint; Fair Model Eval Needs Unified Harness — altryne · 2026-07-30
- Polymarket Odds: 54% Chance of New Gemini Pro Release by Next Month — Polymarket · 2026-07-30
- P-Image-Ideogram Hits Pareto Frontier for Image Gen Speed and Cost — _akhaliq · 2026-07-30