paramjeetn/ragbench-frontend repository overview

⁠RagBench Frontend

The RagBench Frontend is a Next.js 16 web application that provides the interactive dashboard for RagBench — an open-source platform for evaluating, comparing, and optimizing RAG (Retrieval-Augmented Generation) pipelines.

Built with a high-contrast Bauhaus-inspired design system, the frontend provides a complete interface for managing RAG projects, testing pipelines, and analyzing evaluation results.

⁠🚀 Quick Start with Docker Compose

The easiest way to run RagBench is using the pre-built Docker images.

⁠1. Download the Docker Compose file

curl -O https://raw.githubusercontent.com/paramjeetn/RagBench/main/docker-compose.hub.yml

⁠2. Start the complete stack

docker compose -f docker-compose.hub.yml up -d

⁠3. Open the application

The RagBench dashboard will be available at:

http://localhost:3000⁠

The backend API will be available at:

http://localhost:8000⁠

Swagger API documentation:

http://localhost:8000/docs⁠

To stop the stack:

docker compose -f docker-compose.hub.yml down

⁠🎨 Features

⁠Projects

Organize evaluation datasets, configurations, and evaluation runs into separate projects.

⁠Document Studio

Upload documents and inspect the complete ingestion pipeline, including:

• Document parsing
• Chunking
• Embedding
• Vector indexing

⁠Interactive RAG Chat

Test your RAG pipeline interactively and inspect:

• Retrieved documents
• Source citations
• Generated responses
• Retrieval behavior

⁠Evaluation Runner

Run RAG evaluation test suites and visualize metrics including:

• Faithfulness
• Answer Relevancy
• Contextual Precision
• Contextual Recall

⁠Side-by-Side Comparison

Compare two evaluation runs using visual metric comparisons and radar charts to quickly identify differences between pipelines.

⁠Pipeline Configuration

Configure and experiment with:

• Chunking strategies
• Retrieval modes
• Reranking
• LLM providers
• LLM models
• API keys

API keys are stored locally in the browser and are not persisted by the frontend backend.

⁠🏗️ Architecture

The frontend communicates with the RagBench backend API:

┌─────────────────────┐
│   RagBench Frontend │
│     Next.js 16      │
│     :3000           │
└──────────┬──────────┘
           │
           │ REST API
           ▼
┌─────────────────────┐
│   RagBench Backend  │
│      FastAPI        │
│      :8000          │
└──────────┬──────────┘
           │
           ├── Qdrant
           ├── Retrieval
           ├── Reranking
           └── LLM Providers

⁠🔗 Links

• GitHub Repository: https://github.com/paramjeetn/RagBench⁠
• Website & Documentation: https://ragbench-web.vercel.app⁠

⁠📄 License

See the LICENSE⁠ file for licensing information.