Function●Since v1.1

assembleStructuredOutputPipeline

Pipes an LLM through an output parser, optionally wrapping the result to include the raw LLM response alongside the parsed output.

When includeRaw is true, returns { raw: BaseMessage, parsed: RunOutput }. If parsing fails, parsed falls back to null.

assembleStructuredOutputPipeline<
  RunOutput extends Record<string, any> = Record<string, any>
>(
  llm: Runnable<BaseLanguageModelInput>,
  outputParser: Runnable<any, RunOutput>,
  includeRaw: boolean,
  runName: string
): Runnable<BaseLanguageModelInput, RunOutput, RunnableConfig<Record<string, any>>> | Runnable<BaseLanguageModelInput, __type, RunnableConfig<Record<string, any>>>

Parameters

Name	Type	Description
`llm`*	`Runnable<BaseLanguageModelInput>`
`outputParser`*	`Runnable<any, RunOutput>`
`includeRaw`	`boolean`
`runName`	`string`

View source on GitHub

assembleStructuredOutputPipeline

Parameters

LangChain Assistant

Menu

assembleStructuredOutputPipeline

Parameters