jomin7 commented on issue #8864: URL: https://github.com/apache/camel-quarkus/issues/8864#issuecomment-5354619512
Hi @jamesnetherton While implementing the integration tests, I noticed that the `Quarkus LangChain4j embedding model `does not currently return token usage information. The EmbeddingResponse DTO only deserializes the embeddings field, and the result is then wrapped using the single-argument Response.from(embeddings), which leaves tokenUsage as null. As a result, the token count headers in `Camel LangChain4jEmbeddingsProducer` are not populated when using the QL4J integration. It appears that this is specific to the Quarkus fork, as the upstream LangChain4j core handles token usage correctly. For this PR, I’m planning to: Test embedding generation through a Camel route using a Quarkus LangChain4j-managed embedding model. Assert that the token count headers are absent and document this as the current behavior. Would you recommend a different approach, or would it also make sense to file an issue with quarkiverse/quarkus-langchain4j regarding the missing token usage information? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
