Hugging Face's logo
Konstantin Vasserman | Amgix's picture

Konstantin Vasserman | Amgix PRO

amgix
dipankarsarkar's profile picture
·
https://amgix.io

AI & ML interests

Building Amgix (a-MAG-ix) - Open-Source Hybrid Search System

Recent Activity

reacted to tomaarsen's post with 🔥 23 days ago
🚨 I've just published Sentence Transformers v6.0, introducing MultiVectorEncoder: ColBERT-style late interaction models are now a fourth model type, for training, inference, and interpretation, alongside the dense, sparse, and reranker models! Details: Where a regular embedding model compresses a whole text into one vector, a multi-vector model keeps one vector per token and scores query against document with the MaxSim operator. That preserves token-level matching information that a single vector has to average away. It is also the state of the art for visual document retrieval, where a text query is matched against page images directly, charts and tables included, with no OCR step in between. Any PyLate, Stanford ColBERT, or ColPali checkpoint loads straight into the same familiar API: model.encode_query(), model.encode_document(), and model.similarity() just work, whether the documents are texts or page images. Does it help? LightOn trained LateOn (multi-vector) and DenseOn (dense) on the same data with the same 149M ModernBERT backbone, and the multi-vector model wins on 9 of the 13 NanoBEIR datasets: 0.6868 vs 0.6764 mean NDCG@10. The price is a bigger index, and the new HierarchicalTokenPooling module halves it at roughly no retrieval cost. Antoine Chaffin, Raphaël Sourty, and I wrote a blog post walking through multi-vector models in practice: loading the various checkpoint formats, encoding and scoring, plugging them into a search stack, running them on page images, and keeping the index affordable. Check it out if you want to get started, or just point your Agent to the URL: https://huggingface.co/blog/multi-vector-encoder pip install sentence-transformers==6.0.0 Release notes: https://github.com/huggingface/sentence-transformers/releases/tag/v6.0.0
posted an update about 1 month ago
New static multilingual retrieval embedding model: https://huggingface.co/amgix/static-retrieval-multilingual-69m-v1 Distilled from ibm-granite/granite-embedding-97m-multilingual-r2 via Model2Vec. Built this because I couldn't find a static model that was both multilingual and trained for retrieval, existing options are one or the other. Faster than the leading static models, and scores higher on average across RTEB and NanoBEIR multilingual benchmarks. Full benchmark tables and training details on the card. Feedback welcome, especially from anyone with a real multilingual retrieval workload.
published a model about 1 month ago
amgix/static-retrieval-multilingual-69m-v1
View all activity

Organizations

None yet