Combine documents by mapping a chain over them, then combining results.

Hierarchy (view full)




combineDocumentChain: StuffDocumentsChain

Chain to use to combine results of applying llm_chain to documents.

documentVariableName: string = "context"

Variable name in the LLM chain to put the documents in

ensureMapStep: boolean = false

Ensures that the map step is taken regardless of max tokens

inputKey: string = "input_documents"
llmChain: LLMChain<string, any>

LLM Wrapper to use after formatting documents

maxIterations: number = 10

The maximum number of iterations to run through the map

maxTokens: number = 3000

The maximum number of tokens before requiring to do the reduction

returnIntermediateSteps: boolean = false

Return the results of the map steps in the output.

memory?: any



  • Parameters

    • inputs: ChainValues[]
    • Optional config: any[]

    Returns Promise<ChainValues[]>

    ⚠️ Deprecated ⚠️

    Use .batch() instead. Will be removed in 0.2.0.

    This feature is deprecated and will be removed in the future.

    It is not recommended for use.

    Call the chain on all inputs in the list

  • Parameters

    • values: any
    • Optional config: any
    • Optional tags: string[]


    Returns Promise<ChainValues>


    Use .invoke() instead. Will be removed in 0.2.0.

    Run the core logic of this chain and add to output if desired.

    Wraps _call and handles memory.

  • Invoke the chain with the provided input and returns the output.


    • input: ChainValues

      Input values for the chain run.

    • Optional options: any

    Returns Promise<ChainValues>

    Promise that resolves with the output of the chain run.

  • Parameters

    • inputs: Record<string, unknown>
    • outputs: Record<string, unknown>
    • returnOnlyOutputs: boolean = false

    Returns Promise<Record<string, unknown>>

  • Parameters

    • input: any
    • Optional config: any

    Returns Promise<string>


    Use .invoke() instead. Will be removed in 0.2.0.

Generated using TypeDoc