OpenAI Swarm
StaleDescription
OpenAI Swarm is a lightweight multi-agent collaboration framework focused on simplicity and controllability, ideal for learning and prototyping.
Key Features
- Lightweight multi-agent framework — Multi-agent collaboration via Agent and Handoff primitives
- Stateless design — Built entirely on Chat Completions API, no state between calls
- Function calling integration — Agents can call Python functions and decide next steps from results
- Context variables — Support passing and updating context variables between agents
- Streaming output — Streaming responses for intermediate process visibility
- Testability — Lightweight design makes agent collaboration logic easy to test and debug
Use Cases
💡 Learn and understand multi-agent collaboration patterns
💡 Rapidly prototype multi-agent systems
💡 Build agent networks for task distribution and routing
💡 Implement multi-role collaboration like customer service triage and personal shopping
Strengths & Limitations
✅ Strengths
- • High community interest (21.9k stars)
- • Permissive open-source license (MIT)
- • Established track record (2 years in production)
- • Responsive to issues, low backlog
Tags
Categories
Quick Start
pip install git+https://github.com/openai/swarm.git
from swarm import Swarm, Agent
client = Swarm()
def transfer_to_agent_b():
return agent_b
agent_a = Agent(name="Agent A", instructions="You are helpful.", functions=[transfer_to_agent_b])
agent_b = Agent(name="Agent B", instructions="Only speak in Haikus.")
response = client.run(agent=agent_a, messages=[{"role": "user", "content": "Talk to agent B."}])
print(response.messages[-1]["content"])