BGE (BAAI)
https://huggingface.co/BAAI/bge-large-en-v1.5
BAAI (Beijing Academy of AI) BGE family — bge-large/base/small + bge-m3 multilingual + bge-reranker. One of the most widely-used open embedding families.
Active · Verified 2026-10-26 · fresh as of 2026-10-26 · 18 cited facts, 1 with a third-party source · 1 recorded change · Suggest a correction
At a glance
- Type
- BAAI open embedding family vendor fix
- Tier
- T2
- Section
- Embedding & reranker services
- Latest release
- FlagEmbedding v1.3.0 (pip; date not captured) repo fix
- Pricing
- free (open weights, MIT) repo fix
- Funding
- searched not found fix
- Freshness
- 2026-10-26 fix
- Layer
- model-infra repo fix
- Licensing
- open-source vendor fix
- Developer activity
- OpenRank 1.01 (12-mo avg 3.62) · 11 participants — 2026-09 third-party fix
When to use
Optimised for: searched not found
Anti-fit: searched not found
Pros & cons
Pros
searched not found
Cons
searched not found
Claims & capabilities
Top of MTEB benchmark for open embeddings; widely used in Chinese OSS RAG stacks.
Technical surface
- API surface
- searched not found
- Backend storage
- searched not found
- Deployment
- Self-hosted only — open weights on Hugging Face (FlagEmbedding library); also served by third-party inference APIs
- Embedding model
- searched not found
- Multi-tenancy
- searched not found
- MCP
- searched not found
- A2A
- searched not found
- OpenTelemetry
- searched not found
Alternatives to BGE (BAAI) → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other embedding & reranker services in the catalog, ranked by inbound references.
- Voyage AI (MongoDB) T1
Specialised embedding and reranker provider — voyage-3, voyage-multilingual, voyage-finance, voyage-law, voyage-code. Acquired by MongoDB Feb 2025 for $220M.
- Cohere Embed T1
Cohere's hosted embedding models, now Embed 5 (2026-09-30; Pro and Fast tiers): text, image and fused text-plus-image inputs, 100+ languages, 128K-token context, Matryoshka output dimensions from 256 to 2048 with float, int8 and binary formats. Reranker API (Rerank 5).
- Jina AI Embeddings T1
Jina AI's embedding family — jina-embeddings-v3 (multilingual), jina-clip-v2 (text+image), jina-colbert-v2. Open-weights + hosted API + ColBERT-style.
- OpenAI Embeddings T1
OpenAI's hosted embedding API — text-embedding-3-small, text-embedding-3-large. Most widely used commercial embedding API in OSS RAG/memory stacks.
- Hugging Face Inference API (Embeddings) T1
HF's hosted Inference API and Inference Endpoints — serve any model from the Hub including the BGE/GTE/E5 embedding families.
- Sentence Transformers T1
Reimers & Gurevych's library for training sentence embeddings — substrate under most open-weight embedding models. Now under Hugging Face.
Sources (17 cited facts, 1 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is BAAI open embedding family vendor
- What it does & what's distinct BAAI (Beijing Academy of AI) BGE family — bge-large/base/small + bge-m3 multilingual + b… vendor
- Claims / benefits Top of MTEB benchmark for open embeddings; widely used in Chinese OSS RAG stacks. vendor
- Created 2023 repo
- Latest release FlagEmbedding v1.3.0 (pip; date not captured) repo
- License MIT vendor
- GitHub FlagOpen/FlagEmbedding — ~12.2k stars, last README news 2025-03 repo
- Pricing free (open weights, MIT) repo
- Deployment Self-hosted only — open weights on Hugging Face (FlagEmbedding library); also served by … repo
- Founders / pedigree Beijing Academy of Artificial Intelligence (BAAI / Zhiyuan); FlagEmbedding team. vendor
- Code/weights release Active — last commit 2026-04 · 11,664★ repo
- architecture-deployment-locus local repo
- licensing-bucket open-source vendor
- positioning-platform-vs-product platform repo
- taxonomy-three-layer-rollup model-infra repo
- developer-activity-opendigger {"activity":20.33,"as_of":"2026-09","openrank_12mo":3.62,"openrank_current":1.01,"partic… third-party
- sdk-languages python repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-26
- Latest release first recorded as “FlagEmbedding v1.3.0 (pip; date not captured)” source
- Pricing first recorded as “free (open weights, MIT)” source
- Deployment first recorded as “Self-hosted only — open weights on Hugging Face (FlagEmbedding library); also served by third-party inference APIs” source