@antoine_chaffin: It’s only BEIR but there are almost 10 points gap between v2 and LateOn We also have good evidence that the model gener…
Summary
LateOn, a new generation ColBERT model, achieves a nearly 10-point improvement over v2 on BEIR and generalizes well outside BEIR, with the same usage in PyLate.
View Cached Full Text
Cached at: 05/31/26, 02:53 AM
It’s only BEIR but there are almost 10 points gap between v2 and LateOn We also have good evidence that the model generalize very well outside of BEIR
GTE-ModernColBERT was an upgrade LateOn is a whole new generation And all of them have the exact same usage in PyLate https://t.co/qfxH43r9Iv
Omar Khattab (@lateinteraction): 20M downloads / month is a new record for colbertv2
but people should probably migrate from this ancient October 2021 model to the LateOn colbert model from @raphaelsrty @antoine_chaffin et al (@LightOnIO)
Similar Articles
@raphaelsrty: We're releasing LateOn and DenseOn today. Two open retrieval models, 149M parameters each. LateOn (ColBERT, multi-vecto…
Raphael released two open-source retrieval models, LateOn (ColBERT multi-vector) and DenseOn (single-vector), each 149M parameters and outperforming 4× larger models on BEIR.
@antoine_chaffin: Reason-ModernColBERT nearly solved BrowseComp-Plus, smashing SOTA and outperforming models models 54× bigger Not bad fo…
Reason-ModernColBERT achieves near-perfect results on BrowseComp-Plus, surpassing SOTA and models 54× larger, then Agent-ModernColBERT further improves with minimal training.
@AmelieTabatta: ColBERT models continue to embarrass models 54× their sizes , this is why we trust late interaction @LightOnIO . A 1-ye…
The article highlights how ColBERT models, despite being smaller and older, outperform larger models like Qwen3-embed-8B when coupled with late interaction techniques and minimal fine-tuning.
@raphaelsrty: At 140 million parameters, our LateOn model yield strong results Unrelated to LateOn, I'm really excited by what's happ…
The LateOn model with 140M parameters achieves strong results, and the community is excited about advances in multi-vector models including new CPU indexes and multilingual support.
@LightOnIO: Reason-ModernColBERT topped BrowseComp-Plus with just 149M parameters. Now, Agent-ModernColBERT adds ~10% on top. Reach…
LightOn released Agent-ModernColBERT, a 149M parameter open-source retrieval model that achieves performance comparable to GPT-5 combined with Qwen3-Embed-8B by integrating agent reasoning traces into queries.