RunAnywhereModels class - runanywhere_models library

Properties

hashCode → int: The hash code for this object.
no setterinherited
runtimeType → Type: A representation of the runtime type of the object.
no setterinherited

Methods

available() → Future<List<ModelInfo>>: All available models from the C++ registry.
currentLoadedId(ModelCategory category) → Future<String?>: Currently-loaded model id for a given category, or null when nothing is loaded. Lets callers do the "already-loaded?" check without hand-rolling a per-capability switch.
downloadedModels() → Future<ModelListResult>: All downloaded models. Mirrors Swift RunAnywhere.downloadedModels().
getModel(ModelGetRequest request) → Future<ModelGetResult>: Generated-proto registry get-by-id surface.
inferModelFileRole({required String filename, required ModelCategory modality}) → ModelFileRole: Infer the canonical ModelFileRole for a single sidecar filename in a multi-file model. Mirrors Swift's RunAnywhere.inferModelFileRole(filename:modality:) and delegates to the shared commons classifier rac_infer_model_file_role, so the SDK and the C++ model-paths resolver always agree on which file is the primary model, the vision projector (mmproj), tokenizer, vocabulary, etc.
list({ModelQuery? query}) → Future<ModelListResult>: Generated-proto registry list surface.
listDownloaded() → Future<ModelListResult>: Generated-proto downloaded-model registry surface.
loadModel(ModelInfo model) → Future<void>: Polymorphic load entry — dispatches on ModelInfo.category so callers do not hand-roll a per-capability switch.
noSuchMethod(Invocation invocation) → dynamic: Invoked when a nonexistent method or property is accessed.
inherited
queryModels(ModelQuery query) → Future<ModelListResult>: Generated-proto registry query surface.
refreshModelRegistry({bool rescanLocal = true, bool includeRemoteCatalog = false, bool pruneOrphans = false}) → Future<void>: Refresh the model registry. Routes through the commons C ABI rac_model_registry_refresh_proto.
register({String? id, required String name, required String url, required InferenceFramework framework, ModelCategory modality = ModelCategory.MODEL_CATEGORY_LANGUAGE, ModelArtifactType? artifactType, int? memoryRequirement, bool supportsThinking = false, bool supportsLora = false}) → Future<ModelInfo>: Register a single-file remote model with the SDK.
registerArchiveModel({required String archiveUrl, required ArchiveStructure structure, String? id, required String name, required InferenceFramework framework, ModelCategory modality = ModelCategory.MODEL_CATEGORY_LANGUAGE, ArchiveType? archiveType, int? memoryRequirement, bool supportsThinking = false, bool supportsLora = false}) → Future<ModelInfo>: Register an archive-packaged model (tar.gz / tar.bz2 / tar.xz / zip) where the caller needs to specify the on-disk layout.
registerMultiFile({required String id, required String name, required List<ModelFileDescriptor> files, required InferenceFramework framework, ModelCategory modality = ModelCategory.MODEL_CATEGORY_LANGUAGE, int? memoryRequirement, int? contextLength, bool supportsThinking = false, ModelSource source = ModelSource.MODEL_SOURCE_REMOTE}) → Future<ModelInfo>: Register a multi-file model (e.g. embedding model.onnx + vocab.txt).
remove(String modelId) → Future<void>: Remove a model from the C++ registry (called on delete).
resolveModelFilePath(ModelInfo model) → Future<String>: Resolve the primary load target for a generated ModelInfo.
toString() → String: A string representation of this object.
inherited
unloadModel(ModelInfo model) → Future<void>: Polymorphic unload entry — dispatches on ModelInfo.category. Mirrors Swift RunAnywhere.unloadModel(_:) with category routing.
updateDownloadStatus(String modelId, String? localPath) → Future<void>: Update the download status / local path for a model in the C++ registry. Called after a successful generated-proto download completes. download.

Operators

operator ==(Object other) → bool: The equality operator.
inherited

Static Properties

shared → RunAnywhereModels: no setter