# Hugging Face — Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers

- Company: Hugging Face (huggingface.co)
- Announced: 2026-08-18T00:00:00+00:00
- Category: model-update
- Subject: Platform
- Open weights: yes
- Models affected: PyLate checkpoints, Stanford-NLP ColBERT checkpoints, colpali-engine models, lightonai/mLateOn, lightonai/LateOn
- Source: https://huggingface.co/blog/multi-vector-encoder
- Record: https://forck.live/items/4247-multi-vector-late-interaction-embedding-models-with-sentence-transformers

Sentence Transformers library version 6.0 introduces a new model type called MultiVectorEncoder, which implements ColBERT-style late interaction retrieval. This allows token-level matching using the MaxSim operator, preserving more information than dense single-vector embeddings, particularly beneficial for multi-requirement queries and visual document retrieval. The update enables loading of various checkpoint formats including PyLate, Stanford-NLP ColBERT, and colpali-engine models through the same API.

## Evidence

Verbatim from https://huggingface.co/blog/multi-vector-encoder:

> With the v6.0 update, it gains a fourth model type: MultiVectorEncoder , for ColBERT-style late interaction retrieval.

---

Record: https://forck.live/items/4247-multi-vector-late-interaction-embedding-models-with-sentence-transformers
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md