# FlagEmbedding

> Embedding and reranking models for retrieval

BGE models and tools for retrieval-augmented LLMs.

- Source: https://github.com/FlagOpen/FlagEmbedding
- Homepage: http://www.bge-model.com/
- License: MIT
- Language: Python
- Stars: 12210
- Forks: 927
- Contributors: 69
- Last commit: 2026-08-24
- Latest release: v1.4.2 (2026-08-24)
- Purpose: AI & LLM tooling, RAG & vector DBs
- Runs on: Library / SDK
- For: Personal, Enterprise

## Worth score: 69/100

How much DigGitHub recommends it, from activity, adoption, docs, license and security signals.

- Popularity: 20/25
- Momentum: 10/20
- Maintenance: 21/25
- Community: 10/15
- Readiness: 8/15

## Alternatives

- [txtai](https://diggithub.com/neuml/txtai.md): All-in-one framework for semantic search and LLM workflows
- [PageIndex](https://diggithub.com/VectifyAI/PageIndex.md): Vectorless, reasoning-based RAG
- [Haystack](https://diggithub.com/deepset-ai/haystack.md): AI orchestration for production LLM apps
- [Unstructured](https://diggithub.com/Unstructured-IO/unstructured.md): Convert documents to structured data
- [LightRAG](https://diggithub.com/HKUDS/LightRAG.md): Simple and fast graph-based RAG
- [Memvid](https://diggithub.com/memvid/memvid.md): Single-file memory layer for agents
- [TurboVec](https://diggithub.com/RyanCodrai/turbovec.md): Fast vector index in Rust
- [LangChain4j](https://diggithub.com/langchain4j/langchain4j.md): LLM apps on the JVM

---

Source page: https://diggithub.com/FlagOpen/FlagEmbedding
Updated: 2026-10-04