rag-chat and rag-arena
These are complements: RAG Chat provides the prototyping framework for building RAG applications, while RAG Arena evaluates their quality through user feedback, making them useful together in a development workflow.
About rag-chat
upstash/rag-chat
Prototype SDK for RAG development.
Provides out-of-the-box ingestion for websites, PDFs, and other content sources, with built-in vector storage and optional Redis-backed chat history. Supports streaming in Next.js and integrates with multiple LLM providers (OpenAI, Anthropic, GROQ, Mistral, Ollama) plus observability platforms like Helicone and Langsmith. Optional features include rate limiting and a `disableRag` mode for LLM-only chat applications.
About rag-arena
firecrawl/rag-arena
Open-source RAG evaluation through users' feedback
Implements a competitive RAG evaluation framework where multiple retrieval strategies (vector store, graph-based, multi-query, contextual compression) compete to answer the same queries, with user votes updating Elo ratings stored in Supabase. The Next.js frontend interfaces with a Flask Python service that runs LangChain retrievers, while a dynamic routing layer selects which RAG method to use per query. Real-time leaderboards track retriever performance across voting metrics, enabling systematic comparison of different embedding and document chunking approaches.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work