Open Source • MIT Licensed

AI-Powered Document Chat System

Upload documents, chat with AI about their content, and extract insights instantly. Powered by advanced AI models with semantic search and intelligent document processing.

Open-source document chat platform built with Next.js 15, React 19, and TypeScript. Deploy your own instance with optional billing to monetize as a SaaS.

Watch the complete demo and walkthrough

100% Free to Use

Self-Hosted

Privacy-First

No Vendor Lock-In

Use Cases

Built for Every Industry

From knowledge management to customer support, Document Chat System adapts to your needs

Knowledge Management

Build searchable knowledge bases, research libraries, and personal "second brains" with AI-powered recall

Customer Support

Train support teams with product manuals, policies, and FAQs for instant AI-powered answers

Developer Docs

Create interactive API documentation and technical guides where developers can ask questions

Legal & Compliance

Maintain regulatory documents and get instant answers about compliance requirements and policies

Education & Research

Upload textbooks, lecture notes, and research papers for AI-powered tutoring and study assistance

White-Label SaaS

Deploy as your own branded document management platform with built-in billing and multi-tenancy

And many more: HR onboarding, content creation, medical research, property management, and more

Features

Everything You Need for Document AI

A complete, production-ready platform with advanced features for document management, AI chat, vector search, multi-tenancy, and optional monetization.

Core

Multi-Provider AI Chat

Unified interface supporting OpenRouter (100+ models), OpenAI, and ImageRouter. Switch providers seamlessly with intelligent routing, streaming responses, and conversation history.

100+ AI models via OpenRouter
OpenAI GPT-4 Turbo & GPT-3.5
ImageRouter for image generation
Streaming responses with SSE
Context-aware conversations

Processing

Advanced Document Processing

Intelligent extraction and processing for 5+ file formats including PDFs, Office documents, images, and more. Background processing with Inngest for scalability.

PDF, DOCX, TXT, MD, and images
Image OCR and vision AI
Metadata extraction & analysis
Batch processing with queues
Real-time progress tracking

Vector Search & Embeddings

Semantic search powered by Pinecone or pgvector. Find relevant content across thousands of documents with AI-powered similarity matching.

Pinecone serverless indexes
PostgreSQL pgvector support
Hybrid search (semantic + keyword)
Automatic chunking & embedding
Namespace isolation per org

Multi-Tenant

Multi-Tenant Architecture

Enterprise-grade multi-tenancy with complete data isolation, organization management, role-based access control, and team collaboration features.

Organization-based isolation
Role-based permissions (RBAC)
Team member invitations
Per-org resource limits
Shared folders & documents

Infrastructure

Organization & Folder Management

Hierarchical folder structure with drag-and-drop organization, bulk operations, and intelligent document categorization.

Nested folder hierarchies
Drag-and-drop organization
Bulk move & copy operations
Smart folder suggestions
Folder-level permissions

Background Jobs

Background Job Processing

Scalable background processing with Inngest for document processing, embeddings, notifications, and scheduled tasks.

Inngest serverless functions
Retry logic & error handling
Scheduled cron jobs
Event-driven workflows
Job monitoring & logging

And Much More...

Supabase File Storage

Scalable cloud storage with CDN distribution

Redis Caching

Lightning-fast responses with Upstash Redis

Type-Safe APIs

End-to-end TypeScript with Zod validation

Real-Time Updates

Live document sync and chat streaming

Docker Ready

Production Dockerfile with optimization

Edge Optimized

Fast global delivery with edge functions

How It Works

From Upload to AI Conversation in Seconds

Our intelligent pipeline processes documents automatically and makes them instantly searchable

1. Upload Documents

Drag and drop any document - PDFs, Word documents, text files, images, and more. Organize in folders or let AI categorize them automatically.

2. AI Processing

Advanced document processors extract text, metadata, and meaning. Content is chunked, embedded, and indexed for semantic search in real-time.

3. Start Chatting

Ask questions in natural language. AI retrieves relevant content and generates accurate answers with citations to source documents.

Under the Hood

The technology powering intelligent document conversations

Document Processing

• Custom parsers for each file type
• OCR for scanned documents & images
• Metadata extraction & enrichment
• Intelligent chunking algorithms

Vector Embeddings

• OpenAI text-embedding-3-large
• Pinecone serverless indexes
• Semantic similarity matching
• Hybrid search (vector + keyword)

AI Conversation

• RAG (Retrieval Augmented Generation)
• Context window management
• Source attribution & citations
• Streaming token responses

Background Jobs

• Inngest event-driven functions
• Automatic retries & error handling
• Queue-based processing
• Real-time progress updates

Technology

Built with Modern, Production-Ready Stack

Leveraging the latest technologies for performance, scalability, and developer experience

Frontend

React 19
Next.js 15
TypeScript 5
Tailwind CSS
Radix UI
Zustand

Backend

Next.js API Routes
Prisma ORM
PostgreSQL
Clerk Auth
Zod Validation
tRPC

AI & ML

OpenRouter (100+ models)
OpenAI
ImageRouter
Pinecone
pgvector
LangChain

Infrastructure

Supabase
Upstash Redis
Inngest
Docker
Sentry

All dependencies are actively maintained and production-proven at scale

Open Source

Free Forever, Open to Everyone

Document Chat System is MIT licensed. Use it for personal projects, commercial products, or build your own SaaS business. No restrictions, no vendor lock-in.

MIT License

Permissive license allowing commercial use, modification, and distribution

Community Driven

Active development, regular updates, and welcoming community contributions

No Vendor Lock-In

Self-host anywhere, use your own infrastructure, keep full control of your data

Supported File Formats

100+

AI Models Available

99.9%

Uptime SLA

100%

Open Source

Frequently Asked Questions

What AI models can I use?

You can use any model from OpenRouter (100+ models including GPT-4, Claude, Llama, Mistral, and more),OpenAI directly (GPT-4 Turbo, GPT-3.5), or ImageRouter for image generation. The system supports switching between providers seamlessly.

Do I need to pay for AI API usage?

The platform is free and open source. Configure your AI provider API keys to enable chat features. Supports multiple AI providers including OpenRouter, OpenAI, Anthropic, and more.

How do I self-host this?

Deploy to Vercel with our automated setup script that syncs all environment variables in one command. Or use the included production-ready Dockerfile for any platform that supports Docker (AWS, GCP, Azure, DigitalOcean, Railway, Render). Full step-by-step deployment guide with automated scripts available in the README.

Ready to Transform Your Documents?

Start chatting with your documents today. Free, open source, and ready to deploy.

No credit card required • Deploy in minutes • 100% open source