rags
Basic Information
RAGs is a Streamlit application for building retrieval-augmented generation (RAG) pipelines from a data source using natural language. The app walks a user through describing the dataset (currently a single local file or a web page), specifying the task to form the LLM system prompt, and defining RAG parameters such as top-k, chunk size, summarization, embedding model and LLM. A builder agent generates an initial configuration which can be reviewed and edited in a config view, and once created the generated RAG agent is exposed as a chatbot interface that answers questions over the provided data. The project is built on LlamaIndex, is inspired by GPTs, and includes setup guidance using poetry, a .streamlit/secrets.toml file for an OpenAI key, and a launch command that runs the Streamlit home page file.
Links
Categorization
App Details
-
$$$$ -
$$$$