Arize AI Community Icon

Home
Events

Arize AX Releases
Arize AX Support
Arize News
Arize Observe
Discussions
Introductions
Phoenix Support

Powered by Tightknit

Phoenix Support

Arize Phoenix Integration with Langchain4j: Java Observability Guide | Arize AI Community

Phoenix Support

Arize Phoenix Integration with Langchain4j: Java Observability Guide

·May 27, 2025 06:41 PM

If anyone's interested, I've written a working Arize Phoenix integration with Langchain4j to get observability via Opentelemetry in Java. It uses most of the OpenInference spemconv spec

2 comments

· Sorted by Oldest

RunLLM
APP
·
Phoenix supports OpenTelemetry-compatible tracing, and any instrumentation (including custom ones like your Langchain4j integration) that emits traces following the OpenInference semantic conventions will work with Phoenix for LLM observability. This approach is documented as the recommended way to integrate non-Python frameworks or custom SDKs, as long as the traces follow the OpenInference spec and are sent to the Phoenix collector endpoint (e.g., http://localhost:6006/v1/traces) using an OTLP exporter. For reference, see the OpenInference semantic conventions and Phoenix custom spans documentation.If you want to ensure full compatibility and rich UI features in Phoenix, make sure your spans include the key OpenInference attributes (e.g., span kind, input/output messages, model name, token counts, etc.). Would you like more detail on which attributes are most important or how to test your integration?Sources: - Phoenix custom spans documentation
💡 Hint: Mention RunLLM in the thread for followups.
Kevin S.
·
🔒[private user] I'd be happy to, need to figure out internally how we'd go about sharing, probably a maven module that others could include
🙌1