crewAI/docs/streaming.md

Streaming Support in CrewAI

CrewAI now supports real-time streaming output during crew execution, allowing you to see the progress of agents and tasks as they work.

Basic Usage

from crewai import Agent, Task, Crew
from crewai.llm import LLM

def stream_callback(chunk, agent_role, task_description, step_type):
    print(f"[{agent_role}] {step_type}: {chunk}", end="", flush=True)

llm = LLM(model="gpt-4o-mini", stream=True)

agent = Agent(
    role="Writer",
    goal="Write content",
    backstory="You are a skilled writer.",
    llm=llm
)

task = Task(
    description="Write a short story",
    expected_output="A creative story",
    agent=agent
)

crew = Crew(agents=[agent], tasks=[task])

result = crew.kickoff(
    stream=True,
    stream_callback=stream_callback
)

Multi-Agent Streaming

def enhanced_callback(chunk, agent_role, task_description, step_type):
    print(f"[{agent_role}] {task_description[:20]}... - {step_type}: {chunk}")

researcher = Agent(role="Researcher", ...)
writer = Agent(role="Writer", ...)

research_task = Task(description="Research topic", agent=researcher)
write_task = Task(description="Write article", agent=writer, context=[research_task])

crew = Crew(agents=[researcher, writer], tasks=[research_task, write_task])
result = crew.kickoff(stream=True, stream_callback=enhanced_callback)

Stream Callback Parameters

• chunk: The streaming text chunk
• agent_role: Role of the agent producing the chunk
• task_description: Description of the current task
• step_type: Type of step ("agent_thinking", "final_answer", "llm_response")

Events

The streaming system emits CrewStreamChunkEvent, TaskStreamChunkEvent, and AgentStreamChunkEvent that can be handled using the event bus.

Requirements

• Enable streaming on your LLM: LLM(model="...", stream=True)
• Use the stream=True parameter in crew.kickoff()
• Provide a callback function to handle streaming chunks