endee-io/endee

Health v2 · maintenance only

Adoption not part of the score

1303 stars · 1659 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Endee is a high-performance C++ vector database designed to handle up to 1B vectors on a single node, supporting dense, sparse (BM25), and multi-vector retrieval with HNSW-based ANN search, payload filtering, and quantization. It can be self-hosted via Docker or used as a fully-managed serverless cloud, with client integrations for LangChain, LlamaIndex, and CrewAI.

Use cases

• build a RAG pipeline with semantic search over embeddings
• run hybrid search combining dense vectors and BM25 keyword retrieval
• store and query a billion vectors on a single node
• add filtered vector search with metadata to my app
• power a recommendation engine with fast nearest-neighbor lookup
• deploy vector search on edge devices for visual navigation
• fuse results from multiple vector fields with reciprocal rank fusion

When to choose

• you need very high vector throughput and low latency at billion-scale on a single node
• you want hybrid dense + sparse retrieval with filtering in one platform
• you need both a managed cloud and a self-hosted Docker option
• you're building RAG, semantic search, or recommendation systems and want LangChain/LlamaIndex integrations

When to avoid

• you need a mature ecosystem with broad community support and permissive licensing (AGPL-3.0, relatively new project)
• you only need simple full-text search without vectors
• you require multi-node distributed clustering on a budget (cluster tier is enterprise-priced)
• your stack depends on a vector database with wide third-party tooling and long production track record

Facets

service · maturity active

vector-database search-engine rag caching databases artificial-intelligence machine-learning developer-tools self-hosted cloud cpp ann hnsw hybrid-search semantic-search sparse-vectors bm25 quantization edge-ai similarity-search embeddings search retrieval-augmented-generation docker linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="endee-io/endee")