flutter_edge_ai_sqlite 2.0.0

SQLite vector search (sqlite-vec) provider for flutter_edge_ai_rag across native platforms and web.

example/README.md

flutter_edge_ai_sqlite example

flutter_edge_ai_sqlite is an opt-in vector-store provider for flutter_edge_ai_rag that works on every platform: in-SQLite sqlite-vec/vec0 KNN on native (sqlite3 via dart:ffi) and web (package:sqlite3/wasm + a custom sqlite3.wasm). Register it once and open independently owned RAG indexes.

import 'package:flutter/foundation.dart' show kIsWeb;
import 'package:flutter_edge_ai_rag/flutter_edge_ai_rag.dart';
import 'package:flutter_edge_ai_sqlite/flutter_edge_ai_sqlite.dart';
import 'package:path/path.dart' as p;
import 'package:path_provider/path_provider.dart';

Future<void> buildIndex() async {
  final rag = FlutterEdgeAiRag(
    providers: [const SqliteVectorStoreProvider()],
  );
  // An absolute file path on native, a plain name on Web (path_provider has
  // no Web implementation, so it is only called on native).
  final location = kIsWeb
      ? 'knowledge.db'
      : p.join((await getApplicationDocumentsDirectory()).path, 'knowledge.db');
  final index = await rag.open(
    spec: VectorStoreSpec(providerId: 'sqlite', location: location),
    embeddingProfile: EmbeddingProfile(
      id: 'my-embedder-v1',
      dimension: 768,
    ),
  );

  // Add a document with a pre-computed embedding (e.g. from
  // flutter_edge_ai_litertlm or flutter_edge_ai_onnx).
  await index.addVector(
    id: 'doc-1',
    content: 'Flutter Edge AI runs fully on-device.',
    embedding: List<double>.filled(768, 0.0), // your real embedding here
    metadata: '{"lang":"en"}',
  );

  final hits = await index.searchVector(
    embedding: List<double>.filled(768, 0.0), // your real query vector
    topK: 5,
  );
  for (final h in hits) {
    print('${h.id}: ${h.content} (score ${h.similarity})');
  }
  await index.dispose();
}

On web, the custom sqlite3.wasm (with sqlite-vec linked in) is served as a web asset — no CDN <script> is needed; see the package README for the wasm wiring. Native platforms need no setup (sqlite3 bundles its own library; the vec0 extension is bundled via the package's Native Assets hook). A full runnable app wiring every engine and RAG store together lives in the flutter_edge_ai example.