Redis Vector Library
The production Python client for building AI applications on Redis, covering vector index management, hybrid vector and full-text search, semantic caching, and an MCP server for exposing Redis indexes to AI clients.
RedisVL provides schema design, data loading and CRUD operations for Redis vector indexes, similarity search with metadata filters, complex multi-type filtering, and hybrid search that combines semantic and full-text signals. Its AI-specific extensions include semantic caching to reduce LLM cost and increase throughput, an LLM memory layer for agentic context management, semantic routing for query classification, and embedding caching.
Developer utilities include a CLI for index management from the terminal, async support for indexing and search, more than eight embedding provider integrations, rerankers to improve result relevancy, and an MCP server extra that exposes one or more existing Redis indexes to MCP clients. It installs with pip for Python 3.10 or newer, against Redis Cloud, a local Docker container, Redis Enterprise, or Redis Sentinel for high availability.