VT is a production-ready, privacy-focused AI chat application delivering cutting-edge AI capabilities through an extremely generous free tier and a focused premium subscription. Built with modern web technologies and a privacy-first architecture, VT offers advanced AI reasoning, document processing, web search integration, and comprehensive multi-AI provider support.

Featuring Nano Banana - The Conversational image editor! Generate and iteratively edit images through natural conversation, exclusive to VT.

Live at https://vtchat.io.vn

Advanced AI capabilities including document processing, web search, and structured output

• Live Deployment: Successfully running on Fly.io with 2-region setup
• Zero TypeScript Errors: Complete type safety across 50+ files
• 87% Performance Boost: Compilation optimized from 24s to 3s
• Comprehensive Feature Set: All planned features implemented and tested
• Security Hardened: Bot detection, Better Auth, and privacy-first architecture
• Modern Stack: Next.js 15, React 19, TypeScript, Bun ecosystem

• Local-First Storage: All conversations stored in browser IndexedDB
• Zero Server Chat Storage: Complete privacy with per-user data isolation
• Better Auth Integration: 87% performance improvement with secure sessions
• Production Security: Bot detection and comprehensive protection measures

• Nano Banana Conversational Image Editor (New to VT): Conversational image editor! Generate an image, then iteratively edit through natural conversation: "make the cat bigger", "change background to sunset", "add a party hat" - all while preserving edit history
• Premium AI Models (Free with BYOK): Claude Sonnet 4.5, Claude 4 Sonnet/Opus, GPT-4.1, O3/O3 Mini/O4 Mini, O1 Mini/Preview, Gemini 2.5 Pro, DeepSeek R1, Grok 3 - all available to logged-in users with their own API keys
• 9 Free Server Models: Gemini 2.0/2.5 Flash series + OpenRouter models (DeepSeek V3, Qwen3 14B) - no API keys required
• Free Local AI: Run AI models on your own computer with Ollama and LM Studio - completely free, private, and no API costs
• Multi-AI Provider Support: OpenAI, Anthropic, Google, Fireworks, Together AI, xAI, plus local providers (Ollama, LM Studio)
• Intelligent Tool Router (Free): AI-powered semantic routing automatically activates the right tools based on your queries using OpenAI embeddings and pattern matching
• Document Processing (Free): Upload and analyze PDF, DOC, DOCX, TXT, MD files (up to 10MB) - available to all logged-in users
• Structured Output Extraction (Free): AI-powered JSON data extraction from documents - available to all logged-in users
• Chart Visualization (Free): Create interactive bar charts, line graphs, pie charts, and scatter plots - always available and discoverable through smart UI
• Thinking Mode (Free): Complete AI SDK reasoning tokens support with transparent thinking process - available to all logged-in users
• Mathematical Calculator (Free): Advanced functions including trigonometry, logarithms, and arithmetic - always available and discoverable through smart UI
• Web Search Integration (Free): Real-time web search capabilities for current information - always available and discoverable through smart UI

• Local-First Storage: All chat data stored in browser's IndexedDB via Dexie.js
• Zero Server Storage: Conversations never leave the user's device
• Multi-User Isolation: Complete data separation on shared devices
• Security: Secure authentication with Better Auth
• Privacy-Focused Security: Bot detection and secure authentication with Better Auth

Run AI models on your computer for free with complete privacy:

• Complete Local AI Guide - Choose between Ollama and LM Studio
• Ollama Setup Guide - Command-line local AI (5-minute setup)
• LM Studio Setup Guide - GUI local AI (10-minute setup)

• Shadcn UI Design System: Consistent, accessible interface with dark mode support
• Smart Tool Discovery: All tools always visible and clickable with intelligent gated access dialogs
• Seamless Feature Access: Tools remain enabled for discoverability while proper access controls show contextual dialogs
• Responsive Design: Optimized for desktop and mobile with modern Next.js 15 App Router
• 87% Performance Improvement: Faster compilation (24s → 3s) and optimized load times
• Production-Ready: Deployed on Fly.io with 2-region setup and comprehensive monitoring

VT utilizes a Turborepo-managed monorepo structure:

vtchat/
├── apps/
│   └── web/              # Next.js 15 web application (App Router, React 19)
├── packages/
│   ├── actions/          # Server actions (e.g., feedback)
│   ├── ai/               # AI models, providers, tools, semantic router, workflow logic
│   ├── common/           # Shared React components, hooks, context, store
│   ├── orchestrator/     # Workflow engine and task management
│   ├── shared/           # Shared types, constants, configs, utils, logger
│   ├── tailwind-config/  # Shared Tailwind CSS configuration
│   ├── typescript-config/# Shared TypeScript configurations
│   └── ui/               # Base UI components (from Shadcn UI)
├── docs/                 # Documentation and guides
├── memory-bank/          # Project context and evolution tracking
└── scripts/              # Utility scripts (e.g., data sync)

• Framework: Next.js 15 (App Router) with TypeScript
• React: React 19.0.0 (latest stable)
• Styling: Tailwind CSS + Shadcn UI design system
• State Management: Zustand + React Query
• Animations: Framer Motion
• Icons: Lucide React

• Database: Neon PostgreSQL with Drizzle ORM
• Authentication: Better Auth (modern session management)
• Application Security: Bot detection for authentication and secure OAuth integration
• Payment Processing: Creem.io integration
• Local Storage: IndexedDB via Dexie.js
• Deployment: Fly.io (production-ready)

• Runtime: Bun (package manager + JavaScript runtime)
• Monorepo: Turborepo with optimized caching
• Testing: Vitest with Testing Library
• Code Quality: dprint (formatting) + oxlint (comprehensive linting)
• Type Checking: TypeScript with strict configuration
• Performance Monitoring: React Scan for development performance optimization

• AI Providers: OpenAI, Anthropic, Google, Fireworks, Together AI, xAI
• AI SDK: Vercel AI SDK with reasoning tokens support
• Semantic Tool Router: OpenAI embeddings + pattern matching for intelligent tool activation
• Document Processing: Multi-format file analysis
• Chart Generation: Bar charts, line graphs, pie charts, scatter plots
• Web Search: Real-time grounding capabilities

• Docker & Docker Compose: For the easiest setup
• At least one AI API key: OpenAI, Anthropic, or Google Gemini

The easiest way to get VT running locally is with Docker Compose:

git clone https://github.com/vinhnx/vtchat.git
cd vtchat
cp apps/web/.env.example apps/web/.env.local

Edit apps/web/.env.local and add:

• BETTER_AUTH_SECRET (generate with: node -e "console.log(require('crypto').randomBytes(32).toString('hex'))")
• At least one AI API key (e.g., OPENAI_API_KEY=sk-your-key-here)

# Validate setup (optional but recommended)
./validate-setup.sh

# Start the application
docker-compose up --build

• App: http://localhost:3000
• Health Check: http://localhost:3000/api/health

That's it! VT will be running with PostgreSQL, all dependencies, and hot reload enabled.

If you prefer manual setup without Docker:

• Bun v1.1.19+ (JavaScript runtime & package manager)
• PostgreSQL (local or cloud)
• Git

git clone https://github.com/vinhnx/vtchat.git
cd vtchat
bun install

Option A: Local PostgreSQL

# macOS with Homebrew
brew install postgresql@15
brew services start postgresql@15
createdb vtchat_dev

Option B: Docker PostgreSQL

docker run -d --name vtchat-postgres \
  -e POSTGRES_DB=vtchat_dev \
  -e POSTGRES_USER=vtchat \
  -e POSTGRES_PASSWORD=vtchat_password \
  -p 5432:5432 postgres:15-alpine

cp apps/web/.env.example apps/web/.env.local

Edit apps/web/.env.local with your values:

Required:

DATABASE_URL=postgresql://vtchat:vtchat_password@localhost:5432/vtchat_dev
BETTER_AUTH_SECRET=your-32-char-secret-here
NEXT_PUBLIC_BASE_URL=http://localhost:3000
OPENAI_API_KEY=sk-your-openai-key-here

cd apps/web
bun run generate  # Generate database schema

bun dev

# Database
DATABASE_URL=postgresql://user:pass@localhost:5432/vtchat_dev

# Authentication
BETTER_AUTH_SECRET=your-32-char-secret-key
NEXT_PUBLIC_BASE_URL=http://localhost:3000

# AI Provider (choose at least one)
OPENAI_API_KEY=sk-your-key-here
# OR
ANTHROPIC_API_KEY=sk-ant-your-key-here
# OR
GEMINI_API_KEY=your-key-here

# Social Authentication
GITHUB_CLIENT_ID=your-github-client-id
GITHUB_CLIENT_SECRET=your-github-client-secret

# Payment (Creem.io - skip if not needed)
CREEM_API_KEY=creem_test_your-key
CREEM_WEBHOOK_SECRET=your-webhook-secret
CREEM_PRODUCT_ID=your-product-id

# Additional AI Providers
FIREWORKS_API_KEY=fw-your-key-here

# Development
bun dev                    # Start development server
bun dev --turbopack        # With Turbopack (faster)
bun build                  # Production build
bun start                  # Production server

# Database
cd apps/web
bun run generate          # Generate Drizzle schema

# Code Quality
bun run lint              # Lint with oxlint
bun run fmt               # Format with dprint
bun run fmt:check         # Check formatting

# Testing
bun test                  # Run tests
bun test:coverage         # Tests with coverage

Setup validation:

./validate-setup.sh

Reset Docker environment:

docker-compose down -v && docker-compose up --build

Clear caches:

rm -rf node_modules apps/web/node_modules apps/web/.next
bun install

• Port conflicts: Change ports in docker-compose.yml
• Database issues: Check docker-compose logs postgres
• Build failures: Run docker system prune -f then rebuild
• Auth problems: Regenerate BETTER_AUTH_SECRET

Complete Troubleshooting Guide

• Self-Hosting Guide - Complete setup for self-hosters
• Docker Setup Guide - Detailed Docker instructions
• Local Development - Advanced setup options
• Troubleshooting - Common issues & solutions
• Features - Complete feature documentation
• Architecture - System design & tech stack
• Security - Privacy & security implementation