SQLite Cloud

Getting Started with SQLite AI

Overview

SQLite AI is an advanced extension of SQLite designed to bring AI capabilities directly to the edge. It transforms the world’s most deployed database into a powerful, intelligent local database engine capable of powering modern, AI-driven applications, from mobile apps to embedded systems, IoT devices, and robotics.

At the core of the project is a belief that AI should run where the data is — locally, privately, and efficiently — without always relying on the cloud.

Key Extensions

SQLite AI is built around several modular extensions:

  1. SQLite-AI Integrates on-device AI capabilities (e.g., LLM inference, audio transcription with Whisper) directly into SQLite using extensions that wrap C-based AI libraries such as llama.cpp, whisper.cpp, and more.

  2. SQLite-Vector Adds native vector search to SQLite using optimized in-database structures and distance functions (L2, cosine, dot, etc.), with support for multiple vector formats (Float32, Int8, etc.). Ideal for on-device semantic search and AI retrieval.

  3. SQLite-Sync A local-first sync layer that keeps SQLite databases synchronized across devices and users, with support for CRDTs, delta-based replication, and efficient cloud coordination.

  4. SQLite-JS Embed JavaScript directly into SQLite using a lightweight interpreter (e.g., QuickJS), allowing developers to define SQLite functions and logic in JavaScript for increased flexibility and expressiveness.

Features

  • Works offline-first: No need for a server to run advanced queries or AI models.
  • Cross-platform: Runs on iOS, Android, Windows, Linux, macOS, and WebAssembly.
  • Minimal footprint: Optimized for constrained devices.
  • Modular and open: Built in C with clean, embeddable APIs and no heavy dependencies.