Chipper
A web interface, CLI and modular architecture for retrieval-augmented generation, document splitting, web scraping and query workflows, built with Haystack, Ollama, Hugging Face and Elasticsearch, and runnable as a proxy in front of an Ollama instance.
Chipper indexes documents into Elasticsearch for scalable vector storage and retrieval, chunking and splitting them first, and can scrape and index content from web pages. Queries run through RAG pipelines that can be adjusted for model selection, query parameters and system prompts, and it can also transcribe audio to text.
It can sit as a proxy between an Ollama client, such as Enchanted or Open WebUI, and an Ollama instance, adding retrieval capabilities and protecting the Ollama API with API-key or bearer-token authentication. It supports local models through Ollama as well as remote models through the Hugging Face API, and its offline web UI works without an internet connection.
It runs as a fully containerized Docker service, and the project began as a personal tool for local, private RAG work before growing into a general extensible service.