FlagEmbedding and lightweight-embeddings

These are competitors: FlagEmbedding is an established, production-grade embedding framework with extensive community adoption, while lightweight-embeddings is an alternative API service targeting the same use case of generating embeddings with potentially lower resource overhead.

FlagEmbedding
79
Verified
Maintenance 20/25
Adoption 15/25
Maturity 25/25
Community 19/25
Maintenance 6/25
Adoption 6/25
Maturity 9/25
Community 16/25
Stars: 11,395
Forks: 842
Downloads:
Commits (30d): 17
Language: Python
License: MIT
Stars: 15
Forks: 6
Downloads:
Commits (30d): 0
Language: Python
License: MIT
No risk flags
No Package No Dependents

About FlagEmbedding

FlagOpen/FlagEmbedding

Retrieval and Retrieval-augmented LLMs

Provides dense, sparse, and multi-vector embedding models (including BGE-M3 supporting 100+ languages and 8K context) alongside rerankers and multimodal variants for comprehensive semantic search and RAG pipelines. Built on transformer architectures with support for in-context learning, token compression, and unified retrieval methods—integrates seamlessly with vector databases and LLM frameworks via HuggingFace.

About lightweight-embeddings

lh0x00/lightweight-embeddings

LightweightEmbeddings is a fast, free, and unlimited API service for multilingual embeddings and reranking, with support for both text and images and guaranteed uptime.

Scores updated daily from GitHub, PyPI, and npm data. How scores work