Class AiServiceTokenStream

java.lang.Object

dev.langchain4j.service.AiServiceTokenStream

All Implemented Interfaces:

TokenStream

public class AiServiceTokenStream
extends Object
implements TokenStream

Constructor Summary

Constructors

Constructor	Description
AiServiceTokenStream(AiServiceTokenStreamParameters parameters)	Creates a new instance of AiServiceTokenStream with the given parameters.

Method Summary

Modifier and Type	Method	Description
TokenStream	beforeToolExecution(Consumer<BeforeToolExecution> beforeToolExecutionHandler)	The provided consumer will be invoked right before a tool is executed.
TokenStream	ignoreErrors()	All errors during streaming will be ignored (but will be logged with a WARN log level).
TokenStream	onCompleteResponse(Consumer<ChatResponse> completionHandler)	The provided consumer will be invoked when a language model finishes streaming the final chat response, as opposed to the intermediate response (see TokenStream.onIntermediateResponse(Consumer)).
TokenStream	onError(Consumer<Throwable> errorHandler)	The provided consumer will be invoked when an error occurs during streaming.
TokenStream	onIntermediateResponse(Consumer<ChatResponse> intermediateResponseHandler)	The provided consumer will be invoked when a language model finishes streaming the intermediate chat response, as opposed to the final response (see TokenStream.onCompleteResponse(Consumer)).
TokenStream	onPartialResponse(Consumer<String> partialResponseHandler)	The provided consumer will be invoked every time a new partial textual response (usually a single token) from a language model is available.
TokenStream	onPartialResponseWithContext(BiConsumer<PartialResponse, PartialResponseContext> handler)	The provided consumer will be invoked every time a new partial textual response (usually a single token) from a language model is available.
TokenStream	onPartialThinking(Consumer<PartialThinking> partialThinkingHandler)	The provided consumer will be invoked every time a new partial thinking/reasoning text (usually a single token) from a language model is available.
TokenStream	onPartialThinkingWithContext(BiConsumer<PartialThinking, PartialThinkingContext> handler)	The provided consumer will be invoked every time a new partial thinking/reasoning text (usually a single token) from a language model is available.
TokenStream	onPartialToolCall(Consumer<PartialToolCall> partialToolCallHandler)	The provided consumer will be invoked every time a new partial tool call (usually containing a single token of the tool's arguments) from a language model is available.
TokenStream	onPartialToolCallWithContext(BiConsumer<PartialToolCall, PartialToolCallContext> handler)	The provided consumer will be invoked every time a new partial tool call (usually containing a single token of the tool's arguments) from a language model is available.
TokenStream	onRetrieved(Consumer<List<Content>> contentsHandler)	The provided consumer will be invoked if any Contents are retrieved using RetrievalAugmentor.
TokenStream	onToolExecuted(Consumer<ToolExecution> toolExecutionHandler)	The provided consumer will be invoked right after a tool is executed.
TokenStream	onUnmappedRawEvent(Consumer<Object> rawEventHandler)	The provided consumer will be invoked when a provider emits a raw streaming event that is not already exposed through one of the typed callbacks (such as TokenStream.onPartialResponse(Consumer), TokenStream.onPartialThinking(Consumer) or TokenStream.onToolExecuted(Consumer)).
void	start()	Completes the current token stream building and starts processing.

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

AiServiceTokenStream

public AiServiceTokenStream(AiServiceTokenStreamParameters parameters)

Creates a new instance of AiServiceTokenStream with the given parameters.

Parameters:

parameters - the parameters for creating the token stream

Method Details

onPartialResponse

public TokenStream onPartialResponse(Consumer<String> partialResponseHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked every time a new partial textual response (usually a single token) from a language model is available.

Either this or the TokenStream.onPartialResponseWithContext(BiConsumer) callback can be used if you want to consume tokens as soon as they become available.

Specified by:

onPartialResponse in interface TokenStream

Parameters:

partialResponseHandler - lambda that will be invoked when a model generates a new partial textual response

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onPartialResponseWithContext(BiConsumer)

onPartialResponseWithContext

public TokenStream onPartialResponseWithContext(BiConsumer<PartialResponse, PartialResponseContext> handler)

Description copied from interface: TokenStream

The provided consumer will be invoked every time a new partial textual response (usually a single token) from a language model is available.

Either this or the TokenStream.onPartialResponse(Consumer) callback can be used if you want to consume tokens as soon as they become available.

Specified by:

onPartialResponseWithContext in interface TokenStream

Parameters:

handler - lambda that will be invoked when a model generates a new partial textual response

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onPartialResponse(Consumer)

onPartialThinking

public TokenStream onPartialThinking(Consumer<PartialThinking> partialThinkingHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked every time a new partial thinking/reasoning text (usually a single token) from a language model is available.

Either this or the TokenStream.onPartialThinkingWithContext(BiConsumer) callback can be used if you want to consume thinking tokens as soon as they become available.

Specified by:

onPartialThinking in interface TokenStream

Parameters:

partialThinkingHandler - lambda that will be invoked when a model generates a new partial thinking/reasoning text

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onPartialThinkingWithContext(BiConsumer)

onPartialThinkingWithContext

public TokenStream onPartialThinkingWithContext(BiConsumer<PartialThinking, PartialThinkingContext> handler)

Description copied from interface: TokenStream

The provided consumer will be invoked every time a new partial thinking/reasoning text (usually a single token) from a language model is available.

Either this or the TokenStream.onPartialThinking(Consumer) callback can be used if you want to consume thinking tokens as soon as they become available.

Specified by:

onPartialThinkingWithContext in interface TokenStream

Parameters:

handler - lambda that will be invoked when a model generates a new partial thinking/reasoning text

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onPartialThinking(Consumer)

onPartialToolCall

public TokenStream onPartialToolCall(Consumer<PartialToolCall> partialToolCallHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked every time a new partial tool call (usually containing a single token of the tool's arguments) from a language model is available.

Either this or the TokenStream.onPartialToolCallWithContext(BiConsumer) callback can be used if you want to consume partial tool calls as soon as they become available.

Specified by:

onPartialToolCall in interface TokenStream

Parameters:

partialToolCallHandler - lambda that will be invoked when a model generates a new partial tool call

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onPartialToolCallWithContext(BiConsumer)

onPartialToolCallWithContext

public TokenStream onPartialToolCallWithContext(BiConsumer<PartialToolCall, PartialToolCallContext> handler)

Description copied from interface: TokenStream

The provided consumer will be invoked every time a new partial tool call (usually containing a single token of the tool's arguments) from a language model is available.

Either this or the TokenStream.onPartialToolCall(Consumer) callback can be used if you want to consume partial tool calls as soon as they become available.

Specified by:

onPartialToolCallWithContext in interface TokenStream

Parameters:

handler - lambda that will be invoked when a model generates a new partial tool call

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onPartialToolCall(Consumer)

onRetrieved

public TokenStream onRetrieved(Consumer<List<Content>> contentsHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked if any Contents are retrieved using RetrievalAugmentor.

The invocation happens before any call is made to the language model.

Specified by:

onRetrieved in interface TokenStream

Parameters:

contentsHandler - lambda that consumes all retrieved contents

Returns:

token stream instance used to configure or start stream processing

onIntermediateResponse

public TokenStream onIntermediateResponse(Consumer<ChatResponse> intermediateResponseHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked when a language model finishes streaming the intermediate chat response, as opposed to the final response (see TokenStream.onCompleteResponse(Consumer)). Intermediate chat responses contain ToolExecutionRequests, AI service will execute them after returning from this consumer.

Specified by:

onIntermediateResponse in interface TokenStream

Parameters:

intermediateResponseHandler - lambda that consumes intermediate chat responses

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onCompleteResponse(Consumer)

beforeToolExecution

public TokenStream beforeToolExecution(Consumer<BeforeToolExecution> beforeToolExecutionHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked right before a tool is executed.

Specified by:

beforeToolExecution in interface TokenStream

Parameters:

beforeToolExecutionHandler - lambda that consumes BeforeToolExecution

Returns:

token stream instance used to configure or start stream processing

onUnmappedRawEvent

public TokenStream onUnmappedRawEvent(Consumer<Object> rawEventHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked when a provider emits a raw streaming event that is not already exposed through one of the typed callbacks (such as TokenStream.onPartialResponse(Consumer), TokenStream.onPartialThinking(Consumer) or TokenStream.onToolExecuted(Consumer)).

This acts as an escape hatch for provider-specific events that langchain4j does not model, such as server-tool lifecycle events (e.g., OpenAI's web_search_call.in_progress). Events that are already delivered as partial responses, thinking or tool calls are not repeated here.

The event type depends on the provider implementation. Implementations using the dev.langchain4j.http.client.HttpClient abstraction (e.g., OpenAI, Anthropic, Google AI Gemini) typically expose ServerSentEvent; other implementations can expose provider-specific event objects (e.g., the OpenAI official Responses model exposes the SDK's ResponseStreamEvent).

Specified by:

onUnmappedRawEvent in interface TokenStream

Parameters:

rawEventHandler - lambda that consumes raw provider streaming events

Returns:

token stream instance used to configure or start stream processing

onToolExecuted

public TokenStream onToolExecuted(Consumer<ToolExecution> toolExecutionHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked right after a tool is executed.

The invocation happens after the tool method has finished and before any other tool is executed.

Specified by:

onToolExecuted in interface TokenStream

Parameters:

toolExecutionHandler - lambda that consumes ToolExecution

Returns:

token stream instance used to configure or start stream processing

onCompleteResponse

public TokenStream onCompleteResponse(Consumer<ChatResponse> completionHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked when a language model finishes streaming the final chat response, as opposed to the intermediate response (see TokenStream.onIntermediateResponse(Consumer)).

Please note that ChatResponse.tokenUsage() contains aggregate token usage across all calls to the LLM. It is a sum of ChatResponse.tokenUsage()s of all intermediate responses (TokenStream.onIntermediateResponse(Consumer)).

Specified by:

onCompleteResponse in interface TokenStream

Parameters:

completionHandler - lambda that will be invoked when language model finishes streaming

Returns:

token stream instance used to configure or start stream processing

See Also:

• TokenStream.onIntermediateResponse(Consumer)

onError

public TokenStream onError(Consumer<Throwable> errorHandler)

Description copied from interface: TokenStream

The provided consumer will be invoked when an error occurs during streaming.

Specified by:

onError in interface TokenStream

Parameters:

errorHandler - lambda that will be invoked when an error occurs

Returns:

token stream instance used to configure or start stream processing

ignoreErrors

public TokenStream ignoreErrors()

Description copied from interface: TokenStream

All errors during streaming will be ignored (but will be logged with a WARN log level).

Specified by:

ignoreErrors in interface TokenStream

Returns:

token stream instance used to configure or start stream processing

start

public void start()

Description copied from interface: TokenStream

Completes the current token stream building and starts processing.

Will send a request to LLM and start response streaming.

Specified by:

start in interface TokenStream