# Zvec

> Zvec is a lightweight, lightning-fast, in-process vector database by Alibaba. Built for AI applications with simple APIs, powerful indexing, and zero configuration.

Zvec runs entirely in-process — no server, daemon, or external infrastructure required. Just install the package and start indexing and querying vectors. Battle-tested across demanding production workloads within Alibaba Group.

## Docs

### Vector Database

- [Overview](https://zvec.org/en/docs/db/): Introduction, key features, and getting started
- [Quickstart](https://zvec.org/en/docs/db/quickstart/): Install and start searching in minutes
- [AI-Friendly Docs](https://zvec.org/en/docs/db/ai-friendly/): How to feed Zvec docs to AI coding agents
- [Global Configuration](https://zvec.org/en/docs/db/config/): Configure Zvec settings
- [Benchmarks](https://zvec.org/en/docs/db/benchmarks/): Performance benchmarks

#### Concepts

- [Concepts Overview](https://zvec.org/en/docs/db/concepts/): Core concepts of Zvec
- [Data Modeling](https://zvec.org/en/docs/db/concepts/data-modeling/): How Zvec manages vector and scalar data
- [Vector Embedding](https://zvec.org/en/docs/db/concepts/vector-embedding/): Understanding vector representations
- [Vector Index](https://zvec.org/en/docs/db/concepts/vector-index/): Index types overview
  - [Flat Index](https://zvec.org/en/docs/db/concepts/vector-index/flat-index/)
  - [HNSW Index](https://zvec.org/en/docs/db/concepts/vector-index/hnsw-index/)
  - [HNSW-RaBitQ Index](https://zvec.org/en/docs/db/concepts/vector-index/hnsw-rabitq-index/)
  - [IVF Index](https://zvec.org/en/docs/db/concepts/vector-index/ivf-index/)
  - [Quantization](https://zvec.org/en/docs/db/concepts/vector-index/quantization/)
- [Inverted Index](https://zvec.org/en/docs/db/concepts/inverted-index/): Scalar field filtering

#### Collections

- [Collections Overview](https://zvec.org/en/docs/db/collections/): Managing collections in Zvec
- [Create](https://zvec.org/en/docs/db/collections/create/): Create a new collection
  - [Schema](https://zvec.org/en/docs/db/collections/create/schema/): Define collection schema
  - [Options](https://zvec.org/en/docs/db/collections/create/options/): Collection creation options
- [Open](https://zvec.org/en/docs/db/collections/open/): Open an existing collection
- [Inspect](https://zvec.org/en/docs/db/collections/inspect/): Inspect collection metadata
- [Destroy](https://zvec.org/en/docs/db/collections/destroy/): Delete a collection
- [Optimize](https://zvec.org/en/docs/db/collections/optimize/): Optimize collection performance
- [Schema Evolution](https://zvec.org/en/docs/db/collections/schema-evolution/): Modify collection schema

#### Data Operations

- [Data Operations Overview](https://zvec.org/en/docs/db/data-operations/): Working with documents in Zvec
- [Insert](https://zvec.org/en/docs/db/data-operations/insert/): Insert documents
- [Upsert](https://zvec.org/en/docs/db/data-operations/upsert/): Insert or update documents
- [Update](https://zvec.org/en/docs/db/data-operations/update/): Update existing documents
- [Delete](https://zvec.org/en/docs/db/data-operations/delete/): Delete documents
- [Query](https://zvec.org/en/docs/db/data-operations/query/): Query overview
  - [Single Vector](https://zvec.org/en/docs/db/data-operations/query/single-vector/): Single vector search
  - [Multi Vector](https://zvec.org/en/docs/db/data-operations/query/multi-vector/): Multi-vector search
  - [Filter](https://zvec.org/en/docs/db/data-operations/query/filter/): Filtered search
  - [Hybrid](https://zvec.org/en/docs/db/data-operations/query/hybrid/): Hybrid search (dense + sparse)
  - [Group](https://zvec.org/en/docs/db/data-operations/query/group/): Grouped search
- [Fetch](https://zvec.org/en/docs/db/data-operations/fetch/): Fetch documents by ID

#### Building from Source

- [Building Overview](https://zvec.org/en/docs/db/build/): Build Zvec from source
- [Python](https://zvec.org/en/docs/db/build/python/): Build Python package from source
- [Node.js](https://zvec.org/en/docs/db/build/node/): Build Node.js package from source

### AI Integration

- [Overview](https://zvec.org/en/docs/ai/): Embedding models, rerankers, MCP server, and skills
- [Embedding Models](https://zvec.org/en/docs/ai/embedding/): Convert text into vector representations
- [Reranker](https://zvec.org/en/docs/ai/reranker/): Re-score and reorder search results
- [MCP Server](https://zvec.org/en/docs/ai/mcp/): Expose Zvec as a tool for AI agents via MCP
- [Skills](https://zvec.org/en/docs/ai/skills/): Define reusable, agent-friendly operations

## Per-Page Markdown

Every documentation page is available as clean markdown at a predictable URL:

- Pattern: https://zvec.org/mdx/{lang}/docs/{path}.md
- Example: https://zvec.org/mdx/en/docs/db.md
- Example: https://zvec.org/mdx/en/docs/db/quickstart.md

Use these URLs to feed individual pages as context to AI coding agents (Cursor, Copilot, etc.).

## Optional

- [Full documentation](https://zvec.org/llms-full.txt): Complete documentation content in plain text
- [API Reference](https://zvec.org/api-reference/): Python and Node.js API reference