AI Chat Platform - Platform Chat AI Multi-Model Self-Hosted

Platform chat AI lengkap seperti ChatGPT dengan support multi-model (Ollama, OpenAI, Anthropic, Gemini), RAG, streaming real-time, dan agent capabilities yang dapat di-self host.

Fitur Utama

Backend (SELESAI)

• FastAPI backend dengan async/await
• Multi-model LLM support (Ollama, OpenAI, Anthropic, Gemini)
• Streaming real-time dengan Server-Sent Events
• RAG system dengan ChromaDB dan LangChain
• Agent system dengan tools ecosystem
• Authentication JWT + OAuth2
• API key encryption dengan Fernet
• File upload dan processing
• Supabase PostgreSQL database

Frontend (Struktur Siap)

• Next.js 14 dengan App Router
• TypeScript + TailwindCSS
• Shadcn UI components
• Real-time streaming chat interface
• Model management UI
• Settings dan preferences
• File upload dengan drag-drop

Struktur Project

ai-chat-platform/
├── backend/                    # FastAPI Backend (SELESAI)
│   ├── app/
│   │   ├── api/
│   │   │   ├── deps.py        # Dependencies (auth)
│   │   │   └── routes/        # API Routes
│   │   │       ├── auth.py    # Authentication
│   │   │       ├── chat.py    # Chat streaming
│   │   │       ├── models.py  # Model management
│   │   │       └── files.py   # File upload/RAG
│   │   ├── core/
│   │   │   ├── config.py      # Configuration
│   │   │   ├── database.py    # Database connection
│   │   │   └── security.py    # Security utilities
│   │   ├── models/            # SQLAlchemy models
│   │   ├── schemas/           # Pydantic schemas
│   │   ├── services/
│   │   │   ├── llm/           # LLM services
│   │   │   │   ├── base.py           # Base service
│   │   │   │   ├── ollama.py         # Ollama service
│   │   │   │   ├── openai_service.py # OpenAI service
│   │   │   │   ├── anthropic_service.py # Anthropic
│   │   │   │   ├── gemini_service.py # Gemini
│   │   │   │   └── router.py         # Model router
│   │   │   ├── auth_service.py  # Auth service
│   │   │   ├── rag_service.py   # RAG service
│   │   │   └── agent_service.py # Agent service
│   │   └── main.py            # Main application
│   ├── uploads/               # File uploads
│   ├── requirements.txt       # Python dependencies
│   └── .env.example          # Environment template
├── frontend/                  # Next.js Frontend
│   ├── app/                   # Next.js 14 App Router
│   ├── components/            # React components
│   ├── lib/                   # Utilities
│   ├── hooks/                 # Custom hooks
│   └── types/                 # TypeScript types
└── docs/                      # Documentation



## Quick Start

### 1. Generate Security Keys (CRITICAL)
```bash
cd backend

# Generate SECRET_KEY
python -c "import secrets; print('SECRET_KEY='+secrets.token_urlsafe(32))"

# Generate ENCRYPTION_KEY  
python -c "from cryptography.fernet import Fernet; print('ENCRYPTION_KEY='+Fernet.generate_key().decode())"

# Update backend/.env dengan hasil generate di atas

2. Install Ollama (untuk local models)

# Install Ollama
curl -fsSL https://ollama.com/install.sh | sh

# Pull models
ollama pull llama2
ollama pull nomic-embed-text  # untuk RAG embeddings

3. Start Backend

cd backend

# Install dependencies
pip install -r requirements.txt

# Run server
uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

Backend berjalan di: http://localhost:8000 API Docs: http://localhost:8000/docs

4. Test Backend

# Health check
curl http://localhost:8000/health

# Register user
curl -X POST http://localhost:8000/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{"email":"test@example.com","username":"testuser","password":"password123"}'

API Endpoints

Authentication

• POST /api/auth/register - Register user
• POST /api/auth/login - Login
• GET /api/auth/me - Get profile
• PUT /api/auth/me - Update profile

Chat

• POST /api/chat/stream - Stream chat (SSE)
• POST /api/chat/message - Non-streaming chat

Models

• GET /api/models/available - List models
• GET /api/models/validate - Validate connections
• POST /api/models/ollama/pull - Pull Ollama model

Files & RAG

• POST /api/files/upload - Upload file
• POST /api/files/process/{id} - Process document
• GET /api/files/ - List documents
• POST /api/files/query - Query documents

Documentation

• Setup Guide: docs/SETUP.md - Comprehensive setup instructions
• API Documentation: docs/API.md - Complete API reference
• Deployment Guide: DEPLOYMENT.md - Deployment summary
• Interactive API Docs: http://localhost:8000/docs (Swagger UI)

Technology Stack

Backend (PRODUCTION-READY)

• FastAPI (async Python)
• SQLAlchemy + Pydantic
• LangChain + ChromaDB
• Supabase PostgreSQL
• JWT Authentication

LLM Providers

• Ollama (local)
• OpenAI (GPT-4, GPT-3.5)
• Anthropic (Claude 3)
• Google Gemini

License

MIT