FlutterEdgeAiPlugin class abstract
Low-level platform-interface tier (the federated-plugin SPI) for
flutter_edge_ai. It hosts the abstract runtime types (InferenceModel,
InferenceModelSession, InferenceChat, EmbeddingModel,
SpeechRecognizer, SpeechSynthesizer, …) and the platform-dispatch
singleton instance.
App code should use the FlutterEdgeAi static facade instead — it is the
canonical, complete entry point (install / run / introspect / uninstall /
storage). FlutterEdgeAiPlugin.instance is
retained for engine authors, the federated-plugin mock seam, and backward
compatibility; every consumer-facing operation here is also exposed on the
facade.
- Inheritance
-
- Object
- PlatformInterface
- FlutterEdgeAiPlugin
- Implementers
Constructors
Properties
- hashCode → int
-
The hash code for this object.
no setterinherited
- initializedEmbeddingModel → EmbeddingModel?
-
no setter
- initializedModel → InferenceModel?
-
no setter
- modelManager → ModelFileManager
-
no setter
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
createEmbeddingModel(
{PreferredBackend? preferredBackend, bool supportImage = false, bool supportAudio = false}) → Future< EmbeddingModel> -
Returns the embedder for the active embedding model (set by
FlutterEdgeAi.installEmbedder()ormodelManager.ensureModelReadyFromSpec()) — the one already loaded when it serves the request, else a new one. The active model is the only source: explicit paths were removed in 2.3.0. -
createModel(
{required ModelType modelType, ModelFileType fileType = ModelFileType.task, int maxTokens = 1024, PreferredBackend? preferredBackend, PreferredBackend? preferredVisionBackend, PreferredBackend? preferredAudioBackend, List< int> ? loraRanks, int? maxNumImages, bool supportImage = false, bool supportAudio = false, MultimodalOrder? multimodalOrder, bool? enableSpeculativeDecoding, ActivationDataType? activationDataType, int? maxConcurrentSessions}) → Future<InferenceModel> - Creates and returns a new InferenceModel instance.
-
createSttModel(
{String? modelPath, String? tokenizerPath, PreferredBackend? preferredBackend, String? language}) → Future< SpeechRecognizer> - Creates and returns a new SpeechRecognizer instance.
-
createTtsModel(
{PreferredBackend? preferredBackend, String? language}) → Future< SpeechSynthesizer> - Creates and returns a new SpeechSynthesizer for the active TTS model.
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited
Static Properties
- instance ↔ FlutterEdgeAiPlugin
-
getter/setter pair