Instant SDKOne-line installation. No complex configuration or environment variables needed for basic setup.
Multi-Provider LLMClaude, GPT, Gemini, local models—one unified API. Switch providers with an env var, no code changes.
Workflow OrchestrationDAG execution with dynamic planning and auto-recovery. LLM workflows that handle failures gracefully.
MCP NativeExpose APIs as MCP tools automatically. Claude, Cursor, and AI assistants connect directly to your backend.
Real-time StreamingSSE and WebSocket for live AI responses. Stream tokens as they generate, handle long-running tasks.
Built-in RAGDocument parsing, chunking, embeddings, vector search. PDF, DOCX, images with OCR—all built-in.