TAAFT
Free mode
100% free
Freemium
Free Trial
Deals

BGE M3

BGE M3 is a 569M multilingual embedding model designed as a single retrieval backbone for multiple search styles. BAAI says it supports dense retrieval, sparse retrieval, and ColBERT-style multi-vector retrieval at the same time, while covering more than 100 working languages and long-context inputs up to 8192 tokens. The docs also note training coverage of 170+ languages and the use of MCLS to improve long-text embedding without extra fine-tuning.
Text Gen 7
Released: January 30, 2024

Overview

BGE M3 is BAAIโ€™s multilingual retrieval embedding model built around three strengths: multi-functionality, multi-linguality, and multi-granularity. It supports dense retrieval, sparse retrieval, and multi-vector retrieval in one model, works across 100+ languages, and handles inputs up to 8192 tokens.

About Beijing Academy of Artificial Intelligence (BAAI)

Beijing Academy of Artificial Intelligence (BAAI), also known as Zhiyuan Institute, is a Chinese non-profit AI research laboratory established in November 2018. BAAI conducts fundamental AI research, develops open-source models (including WuDao, BGE embeddings, Emu, and RoboBrain), and fosters collaboration between academia and industry. Known for creating the BGE (BAAI General Embedding) series of embedding models used in RAG systems worldwide.

Industry: Research
Company Size: 260
Location: Beijing, Beijing, CN
Website: baai.ac.cn
View Company Profile

Tools using BGE M3

No tools found for this model yet.

Last updated: April 16, 2026
0 AIs selected
Clear selection
#
Name
Task