JamAI Base

JamAI Base is an AI backend for developers building chatbots and applications that use their own documents and structured data. Its spreadsheet-like interface brings prompts, generated data and LLM workflows into tables.

• Generative workflows: Populate table columns with LLM output and connect generation steps into workflows, with streamed responses for web and mobile apps through a REST API.
• Document knowledge: Upload files such as PDFs and DOCX documents to Knowledge Tables, which prepare their contents for retrieval and supply context to other tables.
• Chatbot memory: Chat Tables manage conversation history and context, and draw on Knowledge Tables to answer questions using your content.
• Retrieval controls: Built-in RAG combines keyword, structured and vector search with query rewriting and reranking to select relevant material for model responses.

The platform supports models including OpenAI GPT-4, Anthropic Claude and Meta Llama. It also manages embeddings and rerankers, with SQLite for structured storage and LanceDB for vector data.

JamAI Base is open source under the Apache 2.0 license and can run as a self-hosted service. A hosted cloud offering is also available and requires an account. Self-hosting the backend doesn't make external model services local; using OpenAI or Anthropic models still involves those providers.