Phoenix and OpenInference currently offer strong LLM observability support in Python, with JavaScript support focused mainly on OpenAI, but there are plans to add more LLM frameworks and SDKs for JavaScript. For Java, there is an OpenInference Java SDK with tracing and auto-instrumentation for LangChain4j, and the project is open to contributions for further expansion. There is no explicit mention of Ruby or Go SDKs, but Go applications can emit OpenTelemetry traces to Phoenix using the OpenTelemetry Go SDK and OpenInference semantic conventions; Ruby support is not mentioned in the available documentation or issues聽<https://github.com/Arize-ai/openinference/blob/main/js/README.md|OpenInference JS README>聽<https://github.com/Arize-ai/phoenix/blob/main/docs/section-sdk-api-reference/openinference-sdk/openinference-java.md|OpenInference Java SDK>聽<https://arize.com/docs/phoenix/tracing/how-to-tracing|Phoenix Tracing Docs>聽<https://github.com/Arize-ai/phoenix/issues/7422|Go SDK Issue>.
Would you like more detail on the current state of JavaScript or Java integrations, or guidance on using OpenTelemetry with Go?
Sources:
- <https://github.com/Arize-ai/openinference/blob/main/js/README.md|OpenInference JS README>
- <https://github.com/Arize-ai/phoenix/blob/main/docs/section-sdk-api-reference/openinference-sdk/openinference-java.md|OpenInference Java SDK>
- <https://github.com/Arize-ai/phoenix/issues/7422|Go SDK Issue>
