generateStructured static method

Future<StructuredOutputResult> generateStructured({
  1. required String prompt,
  2. required JSONSchema schema,
  3. LLMGenerationOptions? options,
})

Generate structured output from a typed JSON schema.

Mirrors Swift RunAnywhere.generateStructured(prompt:schema:options:) (RunAnywhere+StructuredOutput.swift:25-39): caller-supplied options (maxTokens, temperature, topP, systemPrompt, …) are forwarded to the underlying LLM through generateWithStructuredOutput; the resulting raw text is then parsed via the commons extraction/validation path.

Implementation

static Future<StructuredOutputResult> generateStructured({
  required String prompt,
  required JSONSchema schema,
  LLMGenerationOptions? options,
}) async {
  if (!DartBridge.isInitialized) throw SDKException.notInitialized();
  final generation = await generateWithStructuredOutput(
    prompt: prompt,
    structuredOutput: StructuredOutputOptionsDefaults.defaults(
      schema: schema,
    ),
    options: options,
  );
  return RunAnywhereLLM.shared.extractStructuredOutput(
    text: generation.text,
    schema: schema,
  );
}