Skip to content

Latest commit

 

History

History
65 lines (45 loc) · 1.98 KB

langchain4j.md

File metadata and controls

65 lines (45 loc) · 1.98 KB

LangChain4j Adapter

AiSupport implementation backed by LangChain4j StreamingChatLanguageModel. When this JAR is on the classpath, @AiEndpoint automatically uses LangChain4j for streaming.

Maven Coordinates

<dependency>
    <groupId>org.atmosphere</groupId>
    <artifactId>atmosphere-langchain4j</artifactId>
    <version>4.0.13</version>
</dependency>

How It Works

Drop the dependency alongside atmosphere-ai and the framework auto-detects it via ServiceLoader:

@AiEndpoint(path = "/ai/chat", systemPrompt = "You are a helpful assistant")
public class MyChat {

    @Prompt
    public void onPrompt(String message, StreamingSession session) {
        session.stream(message);  // uses LangChain4j automatically
    }
}

The LangChain4jAiSupport implementation has priority 100, which takes precedence over the built-in client (priority 0).

Direct Usage

For full control, use LangChain4jStreamingAdapter directly:

var session = StreamingSessions.start(resource);
model.chat(ChatMessage.userMessage(prompt),
    new AtmosphereStreamingResponseHandler(session));

AtmosphereStreamingResponseHandler

Bridges LangChain4j's StreamingChatResponseHandler to Atmosphere's StreamingSession:

LangChain4j Callback StreamingSession Action
onPartialResponse(text) session.send(text)
onCompleteResponse(response) session.complete()
onError(throwable) session.error(message)

Spring Boot Auto-Configuration

AtmosphereLangChain4jAutoConfiguration bridges a Spring-managed StreamingChatLanguageModel bean to the LangChain4jAiSupport SPI automatically.

Samples

See Also