About chatbots and conversations | Instabase AI Hub Documentation

AI Hub supports advanced document analysis using chatbots and conversations.

Chatbots are shareable document analysis apps that can be used to answer questions about a defined knowledge base.
Conversations support single-user, on-demand analysis of uploaded documents.

Both conversations and chatbots let you analyze documents with the help of LLMs, with the key difference being shareability. Chatbots are optimized for broader knowledge sharing and multi-user access, while conversations are ideal for personal, flexible document exploration.

Review the following table for a comparison of features and functionality across chatbots and conversations:

	Chatbots	Conversations
Shareable	Yes, can be published to the Hub and shared.	No, exist only in a personal workspace.
Knowledge base	Chatbot publisher’s uploaded documents. Chatbot users can’t add or remove files from the knowledge base, but can upload up to 5 documents to compare against the knowledge base.	User’s uploaded documents. Files can be added and removed.
Maximum knowledge base size	Same as the origin conversation. Each chatbot user can additionally upload up to 5 comparison files.	About 500 files or 4,000,000 tokens. This limitation is enforced differently between the user interface and the API, see File limitations and processing for details.
Supported models	• Advanced, multistep (automatic selection based on knowledge base size) • Research mode (user-enabled) Model availability varies when interacting with chatbots by API.	• Advanced, multistep (automatic selection based on message scope) • Research mode (user-enabled) Model availability varies when interacting with conversations by API.
API support	Yes, see the queries endpoint.	Yes, see the conversations endpoint for single-document queries and the queries endpoint for multiple-document queries.