Flutter SQLiteAI Extensions Example

February 11, 2026 ยท View on GitHub

A Flutter app demonstrating how to use multiple SQLiteAI extensions together.

Requirements

  • Dart 3.10+ / Flutter 3.38+

Running

As a Flutter app (with UI)

# Generate platform scaffolding (only needed once)
flutter create . --project-name sqliteai_flutter_example

# Install dependencies and run
flutter pub get
flutter run

As a Dart CLI (no UI)

dart pub get
dart run lib/main.dart

Available Extensions

PackageDescription
sqlite_vectorVector similarity search with SIMD-optimized distance functions
sqlite_jsExecute JavaScript directly in SQLite using QuickJS
sqlite_syncReal-time sync with SQLite Cloud
sqlite_aiOn-device LLM inference with llama.cpp
sqlite_agentConnect to MCP servers for AI agent capabilities
sqlite_mcpExpose databases as MCP resources

What it does

  1. Loads all 6 SQLite extensions at startup
  2. sqlite-vector: Vector similarity search with nearest neighbors
  3. sqlite-js: JavaScript evaluation with js_eval()
  4. sqlite-sync: UUID generation with cloudsync_uuid()
  5. sqlite-ai: Version check (requires GGUF model for full features)
  6. sqlite-mcp: Version check (requires MCP server for full features)
  7. sqlite-agent: Version check (requires sqlite-ai + sqlite-mcp setup)

Expected output

=== sqlite-vector ===
Version: 0.9.85
Nearest neighbors:
  id=1, distance=0.0
  id=3, distance=0.1999998688697815

=== sqlite-js ===
Version: 1.2.3
js_eval("1 + 2 + 3") = 6

=== sqlite-sync ===
Version: 0.8.64
cloudsync_uuid() examples:
  01956a8b-1234-7abc-8def-0123456789ab
  01956a8b-1235-7abc-8def-0123456789ab
(Cloud sync requires cloudsync_init() and network config)

=== sqlite-ai ===
Version: 0.7.59
(Requires GGUF model for inference - see API.md)

=== sqlite-mcp ===
Version: 0.1.5
(Requires MCP server connection - see API.md)

=== sqlite-agent ===
Version: 0.1.6
(Requires sqlite-ai + sqlite-mcp setup - see API.md)

Learn more