Sentence Transformers v6.0 Adds MultiVectorEncoder Support
The release treats ColBERT-style late interaction models as a first-class model type for training, inference, and interpretation.
TLDR
Tomaarsen announced the release of Sentence Transformers v6.0. The update adds MultiVectorEncoder as a native model type that supports ColBERT-style late interaction models. These models now sit alongside dense, sparse, and reranker types for training, inference, and interpretation. Hugging Face retweeted the post. Several researchers quoted the thread, including staff from Weaviate and LightOn, noting that the change makes it simpler to train multi-vector models and that projects such as PyLate now build directly on the library.
Combined views
120.9K
15 Sources, first seen ago
Sources
AD 
Andrew Drozdov@mrdrozdov- likes: 6
- replies: 0
- bookmarks: 0
- reposts: 1
RS Raphaël Sourty@raphaelsrtySentence Transformers is the best option to train a multi-vector model from now on ☺️ We choose to build PyLate on top of ST to benefit from all the hard work from @tomaarsen Since our work has been very convincing and impactful, it ended up back in ST, which I'm very proud…
- likes: 37
- replies: 2
- bookmarks: 8
- reposts: 8
AA Aarush@Aarush1003Using the best embedding/retriever models just became easier!! https://twitter.com/tomaarsen/status/2089714037781143733
- likes: 17
- replies: 0
- bookmarks: 3
- reposts: 3
SM Silvio Martinico@SilvioMartinicoGood news for the late interactors 🤓 https://twitter.com/tomaarsen/status/2089714037781143733
- likes: 22
- replies: 0
- bookmarks: 0
- reposts: 5
JH Jeremy Howard@jeremyphowardThe amazingly fast (only 30M!) and extremely accurate @answerdotai ColBERT model is now supported in Sentence Transformers to make it easy to create and query embedding indexes locally directly in Python! 🥳 https://twitter.com/tomaarsen/status/2089714037781143733
- likes: 449
- replies: 11
- bookmarks: 259
- reposts: 48
ME merve@mervenoyannlate interaction models land to Sentence Transformers 🙌🏼 multimodal ones (ColQwen etc) will also move from transformers to ST soon! 🤗 these require a ton of tricks as they're multi vector, so happy to see this https://twitter.com/tomaarsen/status/2089714037781143733
- likes: 73
- replies: 4
- bookmarks: 14
- reposts: 6