Features
Modular LLM components (ChatModel, AgenticModel, Tools, Memory)
Chain, Graph, and Workflow orchestration
Agent Development Kit with ChatModelAgent, Plan-Execute, DeepAgents
Human-in-the-loop via Interrupt/Resume with checkpointing
Streaming programming with CallOption and A2UI Protocol
Observability via Callback, Trace, and Eino Dev visual debugger
Middleware for cross-cutting concerns (Skill, ChatModelAgent)
Multi-agent collaboration and routing
File system access tools (Local, Ark Sandbox)
Document parsing, transformation, embedding pipeline
Timeout, retry, failover, breakpoint execution
Schemeless LLM tool calling (AgenticToolsNode)
Structured output via JSON Schema optimization
Eino Dev IDE plugin for visual orchestration
@ai_fn decorator for LLM-powered Python functions
@ai_classifier decorator for text classification
Structured data extraction via Pydantic models
Agent loops with tool calling and function calling
Built-in streaming (SSE) support
Async-first API for concurrent applications
Rate limiting and automatic retries
LLM response caching for performance
Concurrency control for workload management
CLI tool for monitoring and debugging
SQLite state store for persistence
OpenAI and Anthropic model support
Embeddings generation for semantic search
Local execution with no cloud dependency
Self-hosted deployable as a library
