Features
Ultra-lightweight personal AI agent
Long-horizon task execution (tens to hundreds of steps)
Context coherence maintenance across long tasks
Kernel-standard agent runtime (embeddable)
Predictable token budgets and spend
Open-source under MIT license
Installable via PyPI, UV, or source pip
Multi-language documentation (9 languages)
No external dependencies beyond Python
Steady, fast, and scalable execution
Multi-agent orchestration with handoffs
Sandbox Agents for containerized long-running tasks
Agent-as-tool delegation (v0.15.0+)
Realtime Agents with gpt-realtime-2 voice support (v0.17.6+)
Human-in-the-loop mechanisms
Automatic session history management
Built-in tracing for debugging
Provider-agnostic LLM support (100+ models via LiteLLM)
Redis session support (optional, v0.17.6+)
Instructions, tools, and guardrails configuration
Jupyter notebook compatibility
Supports OpenAI Responses and Chat Completions APIs
