Knowledge
Unlimited knowledge bases — text, PDFs, images, audio, and video sources.
All vectorization — text, image, and video — is processed exclusively through OpenAI, regardless of which LLM your agents use for conversation.