EmbeddedKalanDB class

A KalanDB paired with a TextEmbedder so callers can search by raw text: the query is embedded on-device, then hybrid-searched in one call.

final db = await EmbeddedKalanDB.open(
  dbPath: '.../textbooks.vdb',
  modelPath: '.../model_quantized.onnx',
  vocabPath: '.../vocab.txt',
);
final hits = await db.search('how to solve a quadratic equation');

Properties

db → KalanDB

The underlying database (also usable directly for keyword-only search).

final

embedder → TextEmbedder

The on-device query embedder.

final

hashCode → int

The hash code for this object.

no setter, inherited

runtimeType → Type

A representation of the runtime type of the object.

no setter, inherited

Methods

close() → Future<void>

Closes the database and releases the embedding session.

noSuchMethod(Invocation invocation) → dynamic

Invoked when a nonexistent method or property is accessed.

inherited

search(String text, {int topK = 10, SearchMode mode = SearchMode.hybrid, double alpha = 0.7, MetadataFilter? filter}) → Future<List<SearchResult>>

Embeds text on-device, then runs a search. In hybrid/vector modes the embedding is used as the query vector; text is also passed for BM25.

toString() → String

A string representation of this object.

inherited

Operators

operator ==(Object other) → bool

The equality operator.

inherited

Static Methods

open({required String dbPath, required String modelPath, required String vocabPath, KalanDBConfig? config, int dimension = 384, Pooling pooling = Pooling.cls, String queryInstruction = TextEmbedder.bgeQueryInstruction, int maxSeqLen = 256}) → Future<EmbeddedKalanDB>

Opens the database and loads the embedding model + vocab.