Try LFM • Docs • LEAP • Discord LFM2.5 Embedding 350M We release two new best in class multilingual retrieval models: LFM2.5 Embedding 350M — A dense bi encoder, one vector per document. Smallest, fastest index. LFM2.5 ColBERT 350M — A late interaction model. One vector per token , matched via MaxSim. Higher accuracy and better generalization at the cost of index size. Both models are 350M params and the first bidirectional members of the LFM family, built on LFM2.5 350M Base. They can be used as a drop in replacement for your current RAG pipeline and target fast, cheap, and reliable multilingual / cross lingual search across 11 languages. Find more details about the bidirectional architecture and training recipe in our blog post. 📄 Model details Property LFM2.5 Embedding 350M LFM2.5 ColBERT 350M Type Dense bi encoder (single vector) Late interaction (per token vectors) Total parameters ~354M ~353M Backbone LFM2.5 350M Base + bi directional patches LFM2.5 350M Base + bi directional patches Layers 17 (10 conv + 6 attn + 1 pool) 17 (10 conv + 6 attn + 1 dense) Vocabulary size 65,536 64,402 Output 1024 dim CLS vector 128 dim per token Similarity Cosine MaxSim Training precision BF16…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy