Latest offers Verdent AI: Subscription streak bonus (+10% up to a +20% cap) View all →
← Brands

Model developer / open-source model publisher

BAAI (Beijing Academy of Artificial Intelligence) is a nonprofit AI research institution and the developer and open-source publisher of the BGE embedding and reranking models.

Official website · FlagEmbedding · Hugging Face models

BGE models support RAG, semantic search, knowledge base retrieval and reranking. Original and v1.5 checkpoints are distinct models; aliases do not add duplicate entries.

Search and filter the model directory →

Verified models · 7

BAAI/bge-large-en-v1.5

An English text embedding model for semantic search and retrieval.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-large-en-v1.5
Category
Embedding
Version
v1.5
Languages
en
Input length
512 tokens
Embedding dimensions
1024
License
MIT
Official announcement
2023-09-12
Verified
2026-10-15

Maximum input sequence length in tokens.

Use cases: RAG, semantic search, knowledge base, retrieval.

Official sources and model revision

https://huggingface.co/BAAI/bge-large-en-v1.5
https://huggingface.co/api/models/BAAI/bge-large-en-v1.5
https://huggingface.co/BAAI/bge-large-en-v1.5/raw/d4aa6901d3a41ba39fb536a557fa166f842b0e09/README.md
https://huggingface.co/BAAI/bge-large-en-v1.5/raw/d4aa6901d3a41ba39fb536a557fa166f842b0e09/config.json
https://huggingface.co/BAAI/bge-large-en-v1.5/raw/d4aa6901d3a41ba39fb536a557fa166f842b0e09/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: d4aa6901d3a41ba39fb536a557fa166f842b0e09

BAAI/bge-large-zh-v1.5

A Chinese text embedding model for semantic search and retrieval.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-large-zh-v1.5
Category
Embedding
Version
v1.5
Languages
zh
Input length
512 tokens
Embedding dimensions
1024
License
MIT
Official announcement
2023-09-12
Verified
2026-10-15

Maximum input sequence length in tokens.

Use cases: RAG, semantic search, knowledge base, retrieval.

Official sources and model revision

https://huggingface.co/BAAI/bge-large-zh-v1.5
https://huggingface.co/api/models/BAAI/bge-large-zh-v1.5
https://huggingface.co/BAAI/bge-large-zh-v1.5/raw/79e7739b6ab944e86d6171e44d24c997fc1e0116/README.md
https://huggingface.co/BAAI/bge-large-zh-v1.5/raw/79e7739b6ab944e86d6171e44d24c997fc1e0116/config.json
https://huggingface.co/BAAI/bge-large-zh-v1.5/raw/79e7739b6ab944e86d6171e44d24c997fc1e0116/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: 79e7739b6ab944e86d6171e44d24c997fc1e0116

BAAI/bge-large-en

An English text embedding model for semantic search and retrieval.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-large-en
Category
Embedding
Version
Original release
Languages
en
Input length
512 tokens
Embedding dimensions
1024
License
MIT
Official announcement
2023-08-02
Verified
2026-10-15

Maximum input sequence length in tokens.

Use cases: RAG, semantic search, knowledge base, retrieval.

Official sources and model revision

https://huggingface.co/BAAI/bge-large-en
https://huggingface.co/api/models/BAAI/bge-large-en
https://huggingface.co/BAAI/bge-large-en/raw/abe7d9d814b775ca171121fb03f394dc42974275/README.md
https://huggingface.co/BAAI/bge-large-en/raw/abe7d9d814b775ca171121fb03f394dc42974275/config.json
https://huggingface.co/BAAI/bge-large-en/raw/abe7d9d814b775ca171121fb03f394dc42974275/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: abe7d9d814b775ca171121fb03f394dc42974275

BAAI/bge-large-zh

A Chinese text embedding model for semantic search and retrieval.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-large-zh
Category
Embedding
Version
Original release
Languages
zh
Input length
512 tokens
Embedding dimensions
1024
License
MIT
Official announcement
2023-08-02
Verified
2026-10-15

Maximum input sequence length in tokens.

Use cases: RAG, semantic search, knowledge base, retrieval.

Official sources and model revision

https://huggingface.co/BAAI/bge-large-zh
https://huggingface.co/api/models/BAAI/bge-large-zh
https://huggingface.co/BAAI/bge-large-zh/raw/b5d9f5c027e87b6f0b6fa4b614f8f9cdc45ce0e8/README.md
https://huggingface.co/BAAI/bge-large-zh/raw/b5d9f5c027e87b6f0b6fa4b614f8f9cdc45ce0e8/config.json
https://huggingface.co/BAAI/bge-large-zh/raw/b5d9f5c027e87b6f0b6fa4b614f8f9cdc45ce0e8/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: b5d9f5c027e87b6f0b6fa4b614f8f9cdc45ce0e8

BAAI/bge-m3

A multilingual embedding model supporting dense, sparse and multi-vector retrieval for short text and long documents.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-m3
Category
Embedding
Version
m3
Languages
multilingual
Input length
8,192 tokens
Embedding dimensions
1024
License
MIT
Official announcement
2024-01-30
Verified
2026-10-15

Maximum input sequence length in tokens.

Use cases: RAG, semantic search, knowledge base, retrieval.

Retrieval methods: Dense retrieval, Sparse lexical retrieval, Multi-vector retrieval.

Official sources and model revision

https://huggingface.co/BAAI/bge-m3
https://huggingface.co/api/models/BAAI/bge-m3
https://huggingface.co/BAAI/bge-m3/raw/5617a9f61b028005a4858fdac845db406aefb181/README.md
https://huggingface.co/BAAI/bge-m3/raw/5617a9f61b028005a4858fdac845db406aefb181/config.json
https://huggingface.co/BAAI/bge-m3/raw/5617a9f61b028005a4858fdac845db406aefb181/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: 5617a9f61b028005a4858fdac845db406aefb181

BAAI/bge-reranker-large

A Chinese and English cross-encoder that scores query-document pairs for reranking retrieved results.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-reranker-large
Category
Reranker
Version
Original release
Languages
en, zh
Input length
512 tokens
Embedding dimensions
Not applicable — relevance score output
License
MIT
Official announcement
2023-09-12
Verified
2026-10-15

Maximum tokenized query-document pair length, including special tokens.

Use cases: RAG, reranking, retrieval, knowledge base.

Official sources and model revision

https://huggingface.co/BAAI/bge-reranker-large
https://huggingface.co/api/models/BAAI/bge-reranker-large
https://huggingface.co/BAAI/bge-reranker-large/raw/55611d7bca2a7133960a6d3b71e083071bbfc312/README.md
https://huggingface.co/BAAI/bge-reranker-large/raw/55611d7bca2a7133960a6d3b71e083071bbfc312/config.json
https://huggingface.co/BAAI/bge-reranker-large/raw/55611d7bca2a7133960a6d3b71e083071bbfc312/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: 55611d7bca2a7133960a6d3b71e083071bbfc312

BAAI/bge-reranker-base

A Chinese and English cross-encoder that scores query-document pairs for reranking retrieved results.

No official API pricing published

Open-source model weights can be downloaded and self-hosted under their license. Hosting, compute and third-party API inference may incur separate charges. No verified, directly purchasable API quote is listed here.

Model ID
BAAI/bge-reranker-base
Category
Reranker
Version
Original release
Languages
en, zh
Input length
512 tokens
Embedding dimensions
Not applicable — relevance score output
License
MIT
Official announcement
2023-09-12
Verified
2026-10-15

Maximum tokenized query-document pair length, including special tokens.

Use cases: RAG, reranking, retrieval, knowledge base.

Official sources and model revision

https://huggingface.co/BAAI/bge-reranker-base
https://huggingface.co/api/models/BAAI/bge-reranker-base
https://huggingface.co/BAAI/bge-reranker-base/raw/2cfc18c9415c912f9d8155881c133215df768a70/README.md
https://huggingface.co/BAAI/bge-reranker-base/raw/2cfc18c9415c912f9d8155881c133215df768a70/config.json
https://huggingface.co/BAAI/bge-reranker-base/raw/2cfc18c9415c912f9d8155881c133215df768a70/tokenizer_config.json
https://github.com/FlagOpen/FlagEmbedding

Revision: 2cfc18c9415c912f9d8155881c133215df768a70