Model reference · open weights

bge-m3

Available as managed deployment Embeddings BAAI Embeddings 1 variants 37.1M dl/mo

bge-m3 is an open-weight embedding model from BAAI. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

MakerBAAI
TypeEmbedding models
TaskEmbeddings
Context8194 tokens
Runs withsentence-transformers
Released2024-01-27
Popularity37.1M downloads / month
LicenceOpen weights

About

What bge-m3 is

For more details please refer to our github repo: https://github.com/FlagOpen/FlagEmbedding

BGE-M3 (paper, code)

In this project, we introduce BGE-M3, which is distinguished for its versatility in Multi-Functionality, Multi-Linguality, and Multi-Granularity.

  • Multi-Functionality: It can simultaneously perform the three common retrieval functionalities of embedding model: dense retrieval, multi-vector retrieval, and sparse retrieval.
  • Multi-Linguality: It can support more than 100 working languages.
  • Multi-Granularity: It is able to process inputs of different granularities, spanning from short sentences to long documents of up to 8192 tokens.

Some suggestions for retrieval pipeline in RAG

We recommend to use the following pipeline: hybrid retrieval + re-ranking.

Read the full model card
  • Hybrid retrieval leverages the strengths of various methods, offering higher accuracy and stronger generalization capabilities. A classic example: using both embedding retrieval and the BM25 algorithm. Now, you can try to use BGE-M3, which supports both embedding and sparse retrieval. This allows you to obtain token weights (similar to the BM25) without any additional cost when generate dense embeddings. To use hybrid retrieval, you can refer to Vespa and Milvus.

  • As cross-encoder models, re-ranker demonstrates higher accuracy than bi-encoder embedding model. Utilizing the re-ranking model (e.g., bge-reranker, bge-reranker-v2) after retrieval can further filter the selected text.

News:

  • 2024/7/1: We update the MIRACL evaluation results of BGE-M3. To reproduce the new results, you can refer to: bge-m3_miracl_2cr. We have also updated our paper on arXiv.

    The previous test results were lower because we mistakenly removed the passages that have the same id as the query from the search results. After correcting this mistake, the overall performance of BGE-M3 on MIRACL is higher than the previous results, but the experimental conclusion remains unchanged. The other results are not affected by this mistake. To reproduce the previous lower results, you need to add the --remove-query parameter when using pyserini.search.faiss or pyserini.search.lucene to search the passages.

  • 2024/3/20: Thanks Milvus team! Now you can use hybrid retrieval of bge-m3 in Milvus: pymilvus/examples /hello_hybrid_sparse_dense.py.

  • 2024/3/8: Thanks for the experimental results from @Yannael. In this benchmark, BGE-M3 achieves top performance in both English and other languages, surpassing models such as OpenAI.

  • 2024/3/2: Release unified fine-tuning example and data

  • 2024/2/6: We release the MLDR (a long document retrieval dataset covering 13 languages) and evaluation pipeline.

  • 2024/2/1: Thanks for the excellent tool from Vespa. You can easily use multiple modes of BGE-M3 following this notebook

Specs

  • Model
Model NameDimensionSequence LengthIntroduction
BAAI/bge-m310248192multilingual; unified fine-tuning (dense, sparse, and colbert) from bge-m3-unsupervised
BAAI/bge-m3-unsupervised10248192multilingual; contrastive learning from bge-m3-retromae
BAAI/bge-m3-retromae--8192multilingual; extend the max_length of xlm-roberta to 8192 and further pretrained via retromae
BAAI/bge-large-en-v1.51024512English model
BAAI/bge-base-en-v1.5768512English model
BAAI/bge-small-en-v1.5384512English model
  • Data
DatasetIntroduction
MLDRDocuemtn Retrieval Dataset, covering 13 languages
bge-m3-dataFine-tuning data used by bge-m3

FAQ

1. Introduction for different retrieval methods

  • Dense retrieval: map the text into a single embedding, e.g., DPR, BGE-v1.5
  • Sparse retrieval (lexical matching): a vector of size equal to the vocabulary, with the majority of positions set to zero, calculating a weight only for tokens present in the text. e.g., BM25, unicoil, and [splade](https://arxi

From the published model card. Full card on the HuggingFace links in the sidebar.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys bge-m3 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (bge-m3 below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/embeddings \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"bge-m3","input":"text to embed"}'

Create an account — your API key is available in the console. 5M tokens/month currently included with every new account at launch.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms