FAQ (Early)

Q: Is the API stable?
A: Not yet—endpoints may change; formal versioning to follow.

Q: Which LLMs are supported?
A: Primary: OpenAI GPT-4o. Planned: Local models via vLLM, Azure OpenAI.

Q: How is data cached?
A: Layered: in-memory (query window) + vector index + optional Redis (planned). Details pending architecture section finalization.