Zvec
by Alibaba
Embeds an in-process vector database into applications for dense, sparse, full-text, and hybrid search with durable local storage, via Python, Node.js, Go, Rust, and Dart SDKs.
Skills
Dense And Sparse Vectors
Stores dense and sparse embeddings with multi-vector queries and a choice of index types that scale from in-memory to on-disk.
Full-Text Search
Queries string fields with natural-language or structured keyword expressions, including an N-gram tokenizer for phrase and code search.
Hybrid Search
Combines vector similarity, full-text search, and structured filters in a single query to narrow results to the precise matches.
Durable Local Storage
Persists collections with write-ahead logging so data survives crashes, and lets multiple processes read one collection at once.
Related Agents
LanceDB
Stores and searches multimodal vectors in an embedded database built on the Lance columnar format, running in-process o…
LightRAG
Builds a graph-based retrieval index over documents so agents can answer questions spanning many sources, served throug…
Typesense
Runs typo-tolerant search with built-in vector, hybrid, and semantic modes over a REST API, self-hosted or on Typesense…
PageIndex
Indexes long documents as a reasoning-based hierarchical tree so agents retrieve by navigating structure instead of emb…