Documentation
Home

AI Integration

Multi-provider AI support built with LangChain.

Supported Models

OpenAI (GPT-4, GPT-3.5)Google GeminiAnthropic Claude

The stack abstracts the complexity of different providers, allowing you to switch models with a simple configuration change or let users choose their preferred model.

Core Modules

Chat

Streaming chat interface with history support.

RAG (Retrieval-Augmented Generation)

Upload documents (PDF, TXT) and query them using vector search (ChromaDB).

Image Generation

Generate images from text prompts.

Summarization

Summarize long texts or documents.

Configuration

AI services are located in src/lib/ai. You need to configure the API keys for the providers you intend to use.

env
# AI keys
OPENAI_API_KEY=...
GOOGLE_API_KEY=...
ANTHROPIC_API_KEY=...

Using the API

You can interact with the AI modules via the standardized API routes.

javascript
// Example: Chat API
GET /api/module/chat?message=Hello&conversationId=123

// Example: RAG Query
POST /api/module/rag
Form Data: { action: 'query', question: '...', ... }