Base44 Superagents-AI agent that does it all
Mini toolsmxbai embed large v1
mxbai-embed-large-v1 is Mixedbread’s flagship English embedding model, a 340M-parameter BERT-style encoder that outputs 1024-dimensional vectors for sentences and passages. Trained on over 700M contrastive pairs plus 30M AnglE-loss triplets, it leads MTEB among efficiently sized models and rivals larger proprietary systems. It supports Matryoshka dimensionality reduction and binary or int8 quantization, making it ideal for low-cost semantic search, RAG, clustering, reranking and classification at scale.
New
Text
Gen 7
Overview
mxbai-embed-large-v1 is Mixedbread’s 340M English-only sentence embedding model that maps text into 1024-dim vectors, trained on 700M+ pairs and 30M triplets, delivering state-of-the-art MTEB performance for retrieval, RAG, clustering and classification.
About mixedbread ai
Mixedbread is an applied research lab fundamentally rethinking retrieval. Our mission is to build the memory for AI, ensuring intelligent systems have perfect context to understand the world and interact with it in meaningful ways.
Location:
San Francisco, California, US
Website:
mixedbread.com
Tools using mxbai embed large v1
No tools found for this model yet.
Last updated: February 25, 2026
