Response synthesizer

Defines a base class for response synthesizers used in Gen AI applications.

ResponseSynthesizer(strategy, streamable=True)

Bases: Component

A response synthesizer that uses a strategy to synthesize the response.

Attributes:

Name	Type	Description
strategy	BaseSynthesisStrategy	The strategy used to synthesize the response.
streamable	bool	A flag to indicate whether the synthesized response will be streamed if an event emitter is provided.
preset	PresetFactory	Factory for creating ResponseSynthesizer instances with preset configurations.

Initializes a new instance of the BaseResponseSynthesizer class.

Parameters:

Name	Type	Description	Default
strategy	BaseSynthesisStrategy	The strategy used to synthesize the response.	required
streamable	bool	A flag to indicate whether the synthesized response will be streamed if an event emitter is provided. Defaults to True.	True

map_reduce(map_lm_request_processor=None, reduce_lm_request_processor=None, chunks_repacker=None, extractor_func=None, batch_size=1, max_iterations=1, streamable=True, map_lm_invoker=None, reduce_lm_invoker=None, map_fallback_lms=None, reduce_fallback_lms=None) classmethod

Builds a ResponseSynthesizer backed by MapReduceSynthesisStrategy.

Examples:

map_invoker = build_lm_invoker(model_id="openai/gpt-5-nano")
map_invoker.prompt.build(system_template=..., user_template=...)
reduce_invoker = build_lm_invoker(model_id="openai/gpt-5-nano")
reduce_invoker.prompt.build(system_template=..., user_template=...)

synthesizer = ResponseSynthesizer.map_reduce(
    map_lm_invoker=map_invoker, reduce_lm_invoker=reduce_invoker
)

Parameters:

Name	Type	Description	Default
map_lm_request_processor	LMRequestProcessor | None	The deprecated map-phase request processor. Defaults to None.	None
reduce_lm_request_processor	LMRequestProcessor | None	The deprecated reduce-phase request processor. Defaults to None.	None
chunks_repacker	Repacker | None	The repacker used to batch chunks. Defaults to None.	None
extractor_func	Callable[[LMOutput], Any] | None	The function used to extract the final result from the LM output. Defaults to None.	None
batch_size	int	The number of chunks mapped per batch. Defaults to 1.	1
max_iterations	int	The maximum number of reduce iterations. Defaults to 1.	1
streamable	bool	Whether the synthesizer streams output. Defaults to True.	True
map_lm_invoker	BaseLMInvoker | None	The map-phase invoker to construct this strategy with. Defaults to None.	None
reduce_lm_invoker	BaseLMInvoker | None	The reduce-phase invoker to construct this strategy with. Defaults to None.	None
map_fallback_lms	list[BaseLMInvoker] | None	The fallback invokers to use if map_lm_invoker fails. Defaults to None.	None
reduce_fallback_lms	list[BaseLMInvoker] | None	The fallback invokers to use if reduce_lm_invoker fails. Defaults to None.	None

Returns:

Name	Type	Description
ResponseSynthesizer	ResponseSynthesizer	The constructed synthesizer.

Raises:

Type	Description
ValueError	Validation is delegated to MapReduceSynthesisStrategy.

refine(lm_request_processor=None, batch_size=1, extractor_func=None, stream_drafts=False, streamable=True, lm_invoker=None, fallback_lms=None) classmethod

Builds a ResponseSynthesizer with the refine strategy.

Examples:

lm_invoker = build_lm_invoker(model_id="openai/gpt-5-nano")
lm_invoker.prompt.build(system_template=..., user_template=...)

synthesizer = ResponseSynthesizer.refine(lm_invoker=lm_invoker, batch_size=2)
response = await synthesizer.synthesize(query=query, chunks=chunks)

Parameters:

Name	Type	Description	Default
lm_request_processor	LMRequestProcessor | None	The deprecated request processor to migrate from. Defaults to None.	None
batch_size	int	The number of chunks refined per batch. Defaults to 1.	1
extractor_func	Callable[[LMOutput], Any] | None	The function used to extract the final result from the LM output. Defaults to None.	None
stream_drafts	bool	Whether to stream intermediate drafts. Defaults to False.	False
streamable	bool	Whether the synthesizer streams output. Defaults to True.	True
lm_invoker	BaseLMInvoker | None	The LM invoker to construct this synthesizer with, configured with a prompt builder. Defaults to None.	None
fallback_lms	list[BaseLMInvoker] | None	The fallback invokers to use if lm_invoker fails. Defaults to None.	None

Returns:

Name	Type	Description
ResponseSynthesizer	ResponseSynthesizer	The constructed synthesizer.

Raises:

Type	Description
ValueError	If neither or both of lm_request_processor and lm_invoker are provided (raised by RefineSynthesisStrategy.init).

static_list(format_response_func=None, streamable=True) classmethod

Creates a response synthesizer with the static list strategy.

Examples:

synthesizer = ResponseSynthesizer.static_list()
response = await synthesizer.synthesize(chunks=chunks)

Parameters:

Name	Type	Description	Default
format_response_func	Callable[[list[str]], str] | None	A function that formats a list of context as a response. Defaults to None, in which case the default formatter function will be used.	None
streamable	bool	A flag to indicate whether the synthesized response will be streamed if an event emitter is provided. Defaults to True.	True

Returns:

Name	Type	Description
ResponseSynthesizer	ResponseSynthesizer	A response synthesizer with the static list strategy.

stuff(lm_request_processor=None, chunks_repacker=None, extractor_func=None, streamable=True, lm_invoker=None, fallback_lms=None) classmethod

Builds a ResponseSynthesizer backed by StuffSynthesisStrategy.

Examples:

lm_invoker = build_lm_invoker(model_id="openai/gpt-5-nano")
lm_invoker.prompt.build(system_template=..., user_template=...)

synthesizer = ResponseSynthesizer.stuff(lm_invoker=lm_invoker)
response = await synthesizer.synthesize(query=query, chunks=chunks)

Parameters:

Name	Type	Description	Default
lm_request_processor	LMRequestProcessor | None	The deprecated request processor to migrate from. Defaults to None.	None
chunks_repacker	Repacker | None	The repacker used to combine chunks into a single context string. Defaults to None.	None
extractor_func	Callable[[LMOutput], Any] | None	The function used to extract the final result from the LM output. Defaults to None.	None
streamable	bool	Whether the synthesizer streams output. Defaults to True.	True
lm_invoker	BaseLMInvoker | None	The LM invoker to construct this synthesizer with, configured with a prompt builder. Defaults to None.	None
fallback_lms	list[BaseLMInvoker] | None	The fallback invokers to use if lm_invoker fails. Defaults to None.	None

Returns:

Name	Type	Description
ResponseSynthesizer	ResponseSynthesizer	The constructed synthesizer.

Raises:

Type	Description
ValueError	If neither or both of lm_request_processor and lm_invoker are provided (raised by StuffSynthesisStrategy.init).

synthesize(query=None, chunks=None, history=None, extra_contents=None, hyperparameters=None, event_emitter=None, max_lm_calls=5, **kwargs) async

Synthesizes a response using the assigned strategy.

Parameters:

Name	Type	Description	Default
query	str | None	The input query used to synthesize the response. Defaults to None.	None
chunks	list[Chunk] | None	The list of chunks to be used as context. Defaults to None.	None
history	list[Message] | None	The conversation history to be considered in generating the response. Defaults to None.	None
extra_contents	list[MessageContent] | None	A list of extra contents to be included when generating the response. Defaults to None.	None
hyperparameters	dict[str, Any] | None	The hyperparameters to be passed to the language model. Defaults to None.	None
context_list	list[str] | None	The list of context to be included in the response. Defaults to None.	required
event_emitter	EventEmitter | None	The event emitter for handling events during response synthesis. Defaults to None.	None
max_lm_calls	int	The max number of times the language model can be invoked. Defaults to 5.	5
**kwargs	Any	Additional keyword arguments that will be passed to the strategy.	{}

Returns:

Name	Type	Description
Any	Any	The synthesized response.