---
format: "aidr-story-markdown/v1"
id: "2ee266a2aa4a73b66f5fd411e312749ad99ef24c4c6afc63fccd525efbfc7acd"
canonical_url: "https://aidr.today/2ee266a2?lang=en"
title: "Sentence Transformers v6.0 Adds Multi Vector Support for First Native Late Interaction Models"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-08-18T16:53:23.000Z"
category: "Research"
topics: ["sentence-transformers","embeddings","multimodal","llm"]
source_urls: ["https://huggingnews.com/ai/sentence-transformers-v60-adds-multi-vector-support-for-first-native-lat-5460974a","https://x.com/tomaarsen/status/2089714037781143733","https://x.com/tomaarsen/status/2089714047608381705","https://x.com/tonywu_71/status/2089719653022965981","https://x.com/NielsRogge/status/2089739711296201162","https://x.com/tonywu_71/status/2089718622465806438","https://x.com/lateinteraction/status/2089732602630390252","https://x.com/jeremyphoward/status/2089817361604063739"]
summary: "LightOn previously developed the PyLate library to provide late interaction capabilities that were missing from the Sentence Transformers framework. Version 6.0 now incorporates these functions natively via the MultiVectorEncoder, allowing developers to train, infer and interpret ColBERT-style models alongside dense and sparse embedding types. This architecture replaces traditional bi-encoders, which compute similarity between single vectors per document, with a MaxSim similarity metric that operates on token-level sequences to avoid averaging. The release includes support for ColPali-like vision-language models and architectures from Mixedbread AI and Liquid AI, with existing compatibility for vector databases such as Elastic and Qdrant."
---

# Sentence Transformers v6\.0 Adds Multi Vector Support for First Native Late Interaction Models

> [Open the canonical story](<https://aidr.today/2ee266a2?lang=en>)

**Published:** 2026-08-18T16:53:23.000Z
**Category:** Research
**Topics:** sentence\-transformers, embeddings, multimodal, llm

## Summary

LightOn previously developed the PyLate library to provide late interaction capabilities that were missing from the Sentence Transformers framework\. Version 6\.0 now incorporates these functions natively via the MultiVectorEncoder, allowing developers to train, infer and interpret ColBERT\-style models alongside dense and sparse embedding types\. This architecture replaces traditional bi\-encoders, which compute similarity between single vectors per document, with a MaxSim similarity metric that operates on token\-level sequences to avoid averaging\. The release includes support for ColPali\-like vision\-language models and architectures from Mixedbread AI and Liquid AI, with existing compatibility for vector databases such as Elastic and Qdrant\.

## Sources

- [Story source](<https://huggingnews.com/ai/sentence-transformers-v60-adds-multi-vector-support-for-first-native-lat-5460974a>)
- [Story source](<https://x.com/tomaarsen/status/2089714037781143733>)
- [Supporting source](<https://x.com/tomaarsen/status/2089714047608381705>)
- [Supporting source](<https://x.com/tonywu_71/status/2089719653022965981>)
- [Supporting source](<https://x.com/NielsRogge/status/2089739711296201162>)
- [Supporting source](<https://x.com/tonywu_71/status/2089718622465806438>)
- [Story source](<https://x.com/lateinteraction/status/2089732602630390252>)
- [Story source](<https://x.com/jeremyphoward/status/2089817361604063739>)