LightOn Releases Open Multilingual Retrieval Models with 2.8B Training Pairs
antoine_chaffin · x · 2026-07-30
LightOn has released mDenseOn and mLateOn, 307M-parameter multilingual retrieval models.
- Performance: Achieves state-of-the-art on BEIR and MLDR, while remaining highly competitive on MIRACL and MTEB Code.
- Training Data: Translated validated English pairs into 8 languages, creating a massive 2.8 billion pair cross-lingual corpus.
- Generalization: mLateOn demonstrates strong zero-shot capabilities, remaining competitive on 13 languages unseen during retrieval training (e.g., Chinese, Japanese, Korean).
- Fully Open: Models, datasets, and training code are open-sourced under Apache 2.0.
Related event: LightOn Open-Sources Multilingual Retrieval Models mDenseOn and mLateOn(18 posts)→
More from Models
- Rumor: GLM 5.5 and Kimi K3.1 dropping soon, possibly slowed by anti-distillation blocks — zephyr_z9 · 2026-10-16
- Thrive Holdings benchmarks AI on real tax and IT work across 75+ firms — BorisMPower · 2026-10-16
- A Little Encouragement Makes a Small Local Model Draw a Stunning Pelican SVG — Fancy-Snow7 · 2026-10-16
- Rapid-MLX 0.16.2 brings NVIDIA Nemotron 3.5 up to 120B to Apple Silicon Macs — kuchaev · 2026-10-16
- "End of Chinese OSS" claim pushed back: open models get license almost automatically — eliebakouch · 2026-10-16
- Opus 5.5 still leads a month later, but open-weight decision models are closing in — airesearch12 · 2026-10-16