Struct MapReduceDocumentsChain 

pub struct MapReduceDocumentsChain { /* private fields */ }

MapReduceDocumentsChain

Processes documents in two steps:

1. Map: Calls LLM independently for each document to generate an answer
2. Reduce: Merges all independent answers into a final answer

Implementations

impl MapReduceDocumentsChain

pub fn new<L>(llm: L) -> Self

where L: BaseChatModel + Send + Sync + 'static, L::Error: Into<ProviderError>,

Create a new MapReduceDocumentsChain with the given LLM.

pub fn with_map_prompt(self, template: impl Into<String>) -> Self

Set the map-phase prompt template.

pub fn with_reduce_prompt(self, template: impl Into<String>) -> Self

Set the reduce-phase prompt template.

pub fn with_document_variable(self, name: impl Into<String>) -> Self

Set the document variable name used in the map prompt.

pub fn with_input_key(self, key: impl Into<String>) -> Self

Set the input key.

pub fn with_output_key(self, key: impl Into<String>) -> Self

Set the output key.

pub fn with_name(self, name: impl Into<String>) -> Self

Set the chain name.

pub fn with_verbose(self, verbose: bool) -> Self

Set verbose mode.

pub fn with_map_concurrency(self, limit: usize) -> Self

Cap the number of concurrent map-phase LLM calls (P2-6).

Default is unbounded (all documents mapped in parallel). Setting a limit bounds in-flight requests — useful to respect provider rate limits or bound memory on large document sets.

pub fn build_map_prompt(&self, context: &str, input: &str) -> String

Build the map-phase prompt by substituting the document content and input.

pub fn build_reduce_prompt(&self, summaries: &[String], input: &str) -> String

Build the reduce-phase prompt from the per-document summaries and input.

pub async fn invoke_with_documents( &self, documents: Vec<Document>, input: &str, ) -> Result<String, ChainError>

Invoke with documents and input directly.

Trait Implementations

impl BaseChain for MapReduceDocumentsChain

fn stream<'life0, 'async_trait>( &'life0 self, inputs: HashMap<String, Value>, ) -> Pin<Box<dyn Future<Output = Result<ChainStream, ChainError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Stream execution for MapReduceDocumentsChain.

The map phase runs via invoke (parallel, non-streaming, since reduce needs all summaries) — bounded by map_concurrency when configured. The reduce phase is streamed token by token via stream_chat.

fn input_keys(&self) -> Vec<&str>

Get input keys.

fn output_keys(&self) -> Vec<&str>

Get output keys.

fn invoke<'life0, 'async_trait>( &'life0 self, inputs: HashMap<String, Value>, ) -> Pin<Box<dyn Future<Output = Result<ChainResult, ChainError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Execute the Chain.

fn name(&self) -> &str

Get Chain name.

fn invoke_with_config<'life0, 'async_trait>( &'life0 self, inputs: HashMap<String, Value>, config: Option<RunnableConfig>, ) -> Pin<Box<dyn Future<Output = Result<ChainResult, ChainError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Execute the Chain with a RunnableConfig.

fn stream_with_config<'life0, 'async_trait>( &'life0 self, inputs: HashMap<String, Value>, config: Option<RunnableConfig>, ) -> Pin<Box<dyn Future<Output = Result<ChainStream, ChainError>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Stream execute the Chain with a RunnableConfig.

fn validate_inputs( &self, inputs: &HashMap<String, Value>, ) -> Result<(), ChainError>

Validate inputs.