docs — kitedb

DOCUMENTATION

$ Learn how to build high-performance graph databases with vector search using KiteDB.

Try:bun add @kitedb/core

QUICK_START

./install

Set up KiteDB in your project in under 2 minutes.

./quickstart

Build your first graph database in 5 minutes.

Getting_Started

Introduction

High-performance embedded graph database with vector search

Installation

How to install KiteDB in your project

Quick Start

Build your first graph database in 5 minutes

Guides

Schema Definition

Define type-safe node and edge schemas

Queries & CRUD

Create, read, update, delete operations

Graph Traversal

Navigate relationships in your graph

Vector Search

Semantic similarity search with embeddings

Transactions

ACID transactions and isolation levels

Performance Checklist

Choose the fastest write path and config presets

Concurrency

Multi-threaded access and parallel reads

API_Reference

High-Level API

Drizzle-style fluent API

Low-Level API

Direct database primitives

Vector API

Embedding and similarity search

Benchmarks

Overview

Performance benchmarks overview

Graph Benchmarks

Graph database performance

Vector Benchmarks

Vector search performance

Cross-Language

Bindings performance comparison

Deep_Dive

Architecture

How KiteDB is structured internally

Snapshot + Delta

The core storage model

CSR Format

How edges are stored for fast traversal

Single-File Format

The .kitedb file layout

WAL & Durability

Crash recovery and write-ahead logging

MVCC & Transactions

Concurrent access and isolation

Key Index

Fast node lookups by key

Performance

Why KiteDB is fast

$ Press ⌘K to search docs