OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Evaluation & Observability

Rag Arena

RAG Arena is an open-source Next.js application for comparing responses from retrieval-augmented generation chatbots. It presents multiple answers to a query, collects user votes, then reveals the retriever used for each answer and records retriever results on a Supabase-backed leaderboard.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Rag Arena — OpenViST project profile in AI Evaluation & Observability.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use RAG Arena when you want a user-facing comparison of multiple RAG answers and a vote-based view of retriever performance across queries.

Who it is for

  • Teams evaluating alternative RAG retrievers with feedback from people who compare their answers.
  • Developers adapting an open-source evaluation interface built with Next.js, LangChain and Supabase.

Documented uses

  • Collect votes on several RAG responses to the same query.
  • Reveal which retriever produced each response after voting.
  • Track retriever votes, test counts and Elo ratings in a leaderboard.
  • Ingest articles into a vector store for retrieval.
Before you choose

Requirements & limits

  • Votes and Elo ratings summarize feedback on the configured tests; they are not, by themselves, a universal answer-quality measure.
  • The repository documents a Next.js web application plus a Python service, so local setup includes more than the frontend dependency installation.
  • Graph RAG requires a graph store and an index cache; the README describes building and loading these locally.
  • The documented ingestion path uses LangChain text splitting, OpenAI embeddings and Supabase vector storage.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the web application and prepare its configuration

Install the documented Next.js project dependencies and add the environment-variable names required by its README.

Before you start

  • pnpm
  • The project repository
  • Values for the documented OpenAI, Supabase and Upstash environment variables

Steps

  1. Clone the project repository.

  2. Change into the RAG-arena project directory.

  3. Install dependencies with pnpm i.

  4. Set the README's environment variables for OpenAI, Supabase, Upstash Redis, PRODUCTION and PYTHON_MICRO_SERVER.

Expected outcome

The repository's web dependencies are installed and its documented environment-variable configuration is prepared.

Guide limitations

  • This guide covers dependency installation and configuration only; it does not claim that the application has been started or that the credentials have been validated.
  • The README lists the Python service as a separate component and does not provide its setup in these steps.
Documentation 1 ↗
Source-based practical guide

Install the Python service dependencies

Install the dependencies for the documented Python service.

Before you start

  • Poetry
  • The RAG Arena repository

Steps

  1. Change into the python_service directory.

  2. Run poetry install.

Expected outcome

The Python service dependencies are installed through Poetry.

Guide limitations

  • This setup recipe does not start the Flask server or configure a graph store.
  • The README describes additional graph-store and index preparation for Graph RAG.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
firecrawl/rag-arena · Official homepage title · retained in its source language
How we review sourcesSuggest a correction