Package com.vaadin.flow.component.ai.provider

Interface LLMProvider

All Known Implementing Classes:

LangChain4JLLMProvider, SpringAILLMProvider

public interface LLMProvider

Framework-agnostic interface for Large Language Model (LLM) providers. This interface enables AI-powered components to communicate with LLMs without being tied to a specific implementation. Implementations are responsible for managing conversation memory, handling streaming responses, processing vendor-specific tool annotations, and handling file attachments.

Create an instance by constructing the appropriate implementation directly:

 // Spring AI
 LLMProvider provider = new SpringAILLMProvider(chatModel);

 // LangChain4j
 LLMProvider provider = new LangChain4JLLMProvider(streamingChatModel);
 

Since:

25.1

Author:

Vaadin Ltd.

Nested Class Summary

Nested Classes

Modifier and Type	Interface	Description
static interface	LLMProvider.LLMRequest	Represents a request to the LLM containing all necessary context, configuration, and tools.
static interface	LLMProvider.ToolSpec	A framework-agnostic tool definition that the LLM can invoke.

Method Summary

Modifier and Type	Method	Description
default void	setHistory(List<ChatMessage> history, Map<String,List<AIAttachment>> attachmentsByMessageId)	Restores the provider's conversation memory from a list of chat messages with their associated attachments.
reactor.core.publisher.Flux<String>	stream(LLMProvider.LLMRequest request)	Streams a response from the LLM based on the provided request.

Method Details

stream

reactor.core.publisher.Flux<String> stream(LLMProvider.LLMRequest request)

Streams a response from the LLM based on the provided request. This method returns a reactive stream that emits response tokens as they become available from the LLM. The provider manages conversation history internally, so each call to this method adds to the ongoing conversation context.

Threading: this method is called on the thread that triggers the prompt, and the returned stream is subscribed to on that same thread. An implementation whose LLM call blocks must therefore schedule that call itself — for example with subscribeOn(Schedulers.boundedElastic()) — otherwise it occupies the UI thread and holds the session lock for the whole turn, and nothing the turn produces reaches the browser until it ends. The built-in providers expose this as a setBackgroundExecution(boolean) setting, since running the turn on the request thread is the simpler default when a turn is short.

Callers only consume the stream and do not schedule it, so whether a turn runs in the background is decided entirely by the implementation.

Parameters:

request - the LLM request containing user message, system prompt, attachments, and tools, not null

Returns:

a Flux stream that emits response tokens as strings, never null

Throws:

NullPointerException - if request is null

setHistory

default void setHistory(List<ChatMessage> history,
 Map<String,List<AIAttachment>> attachmentsByMessageId)

Restores the provider's conversation memory from a list of chat messages with their associated attachments. Any existing memory is cleared before the new history is applied.

Providers that support setting chat history should override this method.

This method must not be called while a streaming response is in progress.

Parameters:

history - the list of chat messages to restore, not null

attachmentsByMessageId - a map from ChatMessage.messageId() to the list of attachments for that message, not null

Throws:

NullPointerException - if any argument is null

UnsupportedOperationException - if this provider does not support chat history restoration