Do you need to ask a question?

• I have searched the existing question and discussions and this question is not already answered.
• I believe this is a legitimate question, not just a bug or feature request.

Your Question

Currently, from my understanding, lightrag gets connected to openwebui like an ollama endpoint, meaning the api call used is the /generate one, which as per the documentation bypasses the knowledge base and as such the query results are next to useless unless using /local and even then, they are lower quality than the ones on lightrag-webui. Is there a way to properly connect lightrag to openwebui, so it can function as it should (also have a way to set other settings such as tokens, answer type, etc.)?

Additional Context

No response