RAGs

RAGs lets you create a chatbot that answers questions about a single local file or web page. It's for people who want to describe their task in plain language and review the resulting setup through a visual interface.

• Natural language builder: Describe your data and what you want the chatbot to do, and the builder generates its instructions and retrieval settings.
• Questions and summaries: The chatbot can search your data for relevant passages and use an optional summarization tool to answer broader questions.
• Editable settings: Review and change the generated instructions, retrieval depth, text chunk size, embedding model and language model in the configuration view.
• Model choices: The generated chatbot supports OpenAI, Anthropic, Replicate and local Hugging Face models, with OpenAI or Hugging Face embeddings.

RAGs is an open-source Python app built with Streamlit and LlamaIndex, released under the MIT license. You run the app on your own machine or server, but its default setup uses OpenAI for both the builder and the chatbot, so those model requests go to an external service and require an API key and internet access. Local Hugging Face support applies to the generated chatbot and embeddings; it doesn't make the default builder offline. Each chatbot uses a single local file or web page as its data source.