installModel static method

InferenceInstallationBuilder installModel({
  1. required ModelType modelType,
  2. ModelFileType fileType = ModelFileType.task,
})

Start building an inference model installation

Returns type-safe builder for installing inference models with optional LoRA weights. The model will be automatically set as the active inference model after installation.

Parameters:

  • modelType: Model type (gemmaIt, deepSeek, qwen, etc.) - Required
  • fileType: File type (task or binary) - Defaults to task

Example:

// Install Gemma model
await FlutterGemma.installModel(
  modelType: ModelType.gemmaIt,
)
  .fromNetwork('https://example.com/model.task', token: 'hf_...')
  .withProgress((p) => print('$p%'))
  .install();

// Install DeepSeek with LoRA weights
await FlutterGemma.installModel(
  modelType: ModelType.deepSeek,
)
  .fromNetwork('https://example.com/model.task')
  .withLoraFromNetwork('https://example.com/lora.bin')
  .install();

Implementation

static InferenceInstallationBuilder installModel({
  required ModelType modelType,
  ModelFileType fileType = ModelFileType.task,
}) {
  return InferenceInstallationBuilder(
    modelType: modelType,
    fileType: fileType,
  );
}