NyRAG
A no-code tool for building RAG applications by crawling websites or processing documents, then deploying to Vespa for hybrid search with an integrated chat UI and support for any OpenAI-compatible LLM.
NyRAG answers a question through a multi-stage retrieval pipeline: an LLM generates additional search queries to widen retrieval coverage, each query is embedded with a configured SentenceTransformer model, Vespa's nearestNeighbor search retrieves the most relevant chunks using a best_chunk_score ranking profile, results from all queries are aggregated, deduplicated and ranked, and the final context is sent to an LLM to generate a grounded answer.
It works with any OpenAI-compatible API, including OpenRouter, Ollama, LM Studio, vLLM, LocalAI or OpenAI itself, and is used mainly through its own web UI, which manages the lifecycle from configuring a data source, a web crawl or document set, through processing to chatting. It runs locally with Docker, or against a Vespa Cloud account, and installs via pip or uv.