Skip to main content
The Team class enables multiple agents to work together on complex tasks. Teams can operate in different modes, delegate work to members, and coordinate responses.

​
Constructor

from agno import Agent, Team

researcher = Agent(name="Researcher", role="Research")
writer = Agent(name="Writer", role="Writing")

team = Team(
    members=[researcher, writer],
    model="gpt-4o",
    name="Research Team"
)

​
Core Parameters

​
members
List[Agent | Team] | Callable
required
List of agents or teams that make up this team. Can also be a callable factory.
​
model
Model | str
default:"None"
Language model for the team leader to use for coordination.
​
name
str
default:"None"
A descriptive name for the team.
​
id
str
default:"None"
Unique identifier for the team. Auto-generated if not provided.
​
role
str
default:"None"
If this team is part of another team, this is its role.

​
Execution Modes

​
mode
TeamMode
default:"None"
Team execution mode. When set, overrides the boolean flags below.
​
respond_directly
bool
default:"False"
If True, team leader won’t process member responses and returns them directly.
​
delegate_to_all_members
bool
default:"False"
If True, delegates the task to all members instead of deciding which to use.
​
determine_input_for_members
bool
default:"True"
If False, sends the run input directly to member agents.
​
max_iterations
int
default:"10"
Maximum number of iterations for autonomous task loop (mode=tasks).

​
Session & State

​
session_id
str
default:"None"
Default session ID for the team.
​
session_state
Dict[str, Any]
default:"None"
Session state stored in the database.
​
add_session_state_to_context
bool
default:"False"
If True, adds session_state to the context.
​
enable_agentic_state
bool
default:"False"
If True, gives the team tools to update session_state.

​
History & Memory

​
add_team_history_to_members
bool
default:"False"
Send team-level history to members.
​
num_team_history_runs
int
default:"3"
Number of historical runs to include in messages to members.
​
share_member_interactions
bool
default:"False"
If True, sends all member interactions during current run to delegated members.
​
add_history_to_context
bool
default:"False"
Adds messages from chat history to the context.
​
num_history_runs
int
default:"None"
Number of historical runs to include.

​
Database

​
db
BaseDb | AsyncBaseDb
default:"None"
Database for storing team sessions and history.
​
memory_manager
MemoryManager
default:"None"
Memory manager for the team.

​
Knowledge & Tools

​
knowledge
KnowledgeProtocol | Callable
default:"None"
Knowledge base for the team.
​
tools
List[Toolkit | Callable | Function | Dict] | Callable
default:"None"
Tools available to the team leader.
​
get_member_information_tool
bool
default:"False"
If True, adds a tool to get information about team members.
​
read_chat_history
bool
default:"False"
If True, adds a tool to read the chat history.

​
Response Configuration

​
output_schema
Type[BaseModel] | Dict[str, Any]
default:"None"
Pydantic model or JSON schema for structured output.
​
stream
bool
default:"None"
Stream the response from the team.
​
stream_events
bool
default:"None"
Stream intermediate steps.
​
stream_member_events
bool
default:"True"
Stream events from member agents.
​
store_member_responses
bool
default:"False"
Store member agent runs inside the team’s RunOutput.

​
Methods

​
run()

Run the team with a task. 
response = team.run("Research AI trends and write a report")
print(response.content)
Parameters:
  • input (str | List | Dict | Message | BaseModel): The input task
  • stream (bool): Whether to stream the response
  • session_id (str): Optional session ID
  • user_id (str): Optional user ID
Returns: TeamRunOutput or Iterator of events if streaming

​
arun()

Async version of run(). 
response = await team.arun("Research and write")
Run the team and print the response to console. 
team.print_response("Create a marketing plan")

​
save()

Save the team configuration. 
team.save(db=db, stage="published")

​
load()

Load a team from the database. 
team = Team.load(id="my_team", db=db)

​
Example Usage

from agno import Agent, Team

researcher = Agent(
    name="Researcher",
    role="Find information",
    instructions="Search and gather information"
)

writer = Agent(
    name="Writer",
    role="Write content",
    instructions="Write clear, engaging content"
)

team = Team(
    members=[researcher, writer],
    model="gpt-4o",
    instructions="Coordinate research and writing tasks"
)

response = team.run("Write about quantum computing")

Build docs developers (and LLMs) love

Get started for freeTalk to us