openinference

LangChain4j OpenInference Example with Phoenix

This example demonstrates how to use OpenInference instrumentation with LangChain4j and export traces to Arize Phoenix.

Prerequisites

1. Java 11 or higher
2. OpenAI API key
3. (Optional) Phoenix API key if using auth

Running the Example

Step 1: Start Phoenix

Phoenix is an open-source observability platform for LLM applications. Start it using Docker:

docker run -p 6006:6006 -p 4317:4317 arizephoenix/phoenix:latest

This command:

• Exposes port 6006 for the Phoenix web UI
• Exposes port 4317 for the OTLP gRPC endpoint (where traces are sent)

Once started, you can access Phoenix at http://localhost:6006

Step 2: Set your OpenAI API Key

export OPENAI_API_KEY=your-api-key-here

Step 3: Run the Example

From the java directory:

./gradlew :examples:langchain4j-example:run

Step 4: View Traces in Phoenix

1. Open http://localhost:6006 in your browser
2. You should see the traces from your LangChain4j application
3. Click on a trace to see detailed information including:
   • LLM model name
   • Input/output messages
   • Token counts
   • Request parameters (temperature, max_tokens, etc.)
   • Latency information

What’s Being Traced

The example traces:

• LLM calls to OpenAI
• Input prompts and output responses
• Model parameters (temperature, max tokens)
• Token usage (when available)
• Timing information

This site is open source. Improve this page.