Open lordofthejars opened 2 months ago
cc @jmartisk
There are some ways of doing it like https://github.com/bsorrentino/langgraph4j/blob/main/adaptive-rag/src/main/java/dev/langchain4j/adaptiverag/ChromaStore.java, so basically you call everything manually, and inject them manually too: https://github.com/bsorrentino/langgraph4j/blob/main/adaptive-rag/src/main/java/dev/langchain4j/adaptiverag/RetrievalGrader.java#L61
So there is a way to do it, but maybe we could somehow automate this by creating some kind of bean with requestscoped.
Hi, when you try to implement adaptive RAG (https://langchain-ai.github.io/langgraph/tutorials/rag/langgraph_adaptive_rag/) you might need to reuse the retrieved documents from the RAG system and use it in several AI services.
For example, one needs to grade the response, another needs to find hallucinations, etc.
What would be the best option to implement it in quarkus-langchain4j?