ai-mirow-projects
/
swarms
mirror of https://github.com/kyegomez/swarms
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'kyegomez:master' into JSON-Output-Support-Agent

Browse Source
pull/615/head
Sambhav Dixit 2 months ago committed by GitHub
parent a63541c619 2387096ac3
commit 247ecc65e7
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
10 changed files with 317 additions and 219 deletions
Show Stats Download Patch File Download Diff File

1
changelog.md
Unescape View File

@ -0,0 +1 @@
# 5.8.7

2
docs/requirements.txt
Unescape View File

@ -26,7 +26,7 @@ jinja2~=3.1
markdown~=3.7 markdown~=3.7
mkdocs-material-extensions~=1.3 mkdocs-material-extensions~=1.3
pygments~=2.18 pygments~=2.18
pymdown-extensions~=10.10 pymdown-extensions~=10.11
# Requirements for plugins # Requirements for plugins
babel~=2.16 babel~=2.16

9
example.py
Unescape View File

@ -33,11 +33,14 @@ agent = Agent(
context_length=200000, context_length=200000,
return_step_meta=False, return_step_meta=False,
# output_type="json", # output_type="json",
output_type="string", output_type="json", # "json", "dict", "csv" OR "string" soon "yaml" and
streaming_on=False, streaming_on=False,
# auto_generate_prompt=True,
) )
agent.run( print(
"How can I establish a ROTH IRA to buy stocks and get a tax break? What are the criteria" agent.run(
"How can I establish a ROTH IRA to buy stocks and get a tax break? What are the criteria"
)
) )

28
new_prompt.py
Unescape View File

@ -0,0 +1,28 @@
from swarms import Prompt
from swarm_models import OpenAIChat
import os
model = OpenAIChat(
api_key=os.getenv("OPENAI_API_KEY"), model_name="gpt-4o-mini", temperature=0.1
)
# Aggregator system prompt
prompt_generator_sys_prompt = Prompt(
name="prompt-generator-sys-prompt-o1",
description="Generate the most reliable prompt for a specific problem",
content="""
Your purpose is to craft extremely reliable and production-grade system prompts for other agents.
# Instructions
- Understand the prompt required for the agent.
- Utilize a combination of the most effective prompting strategies available, including chain of thought, many shot, few shot, and instructions-examples-constraints.
- Craft the prompt by blending the most suitable prompting strategies.
- Ensure the prompt is production-grade ready and educates the agent on how to reason and why to reason in that manner.
- Provide constraints if necessary and as needed.
- The system prompt should be extensive and cover a vast array of potential scenarios to specialize the agent.
""",
auto_generate_prompt=True,
llm=model,
)
print(prompt_generator_sys_prompt.get_prompt())

4
pyproject.toml
Unescape View File

@ -5,7 +5,7 @@ build-backend = "poetry.core.masonry.api"
[tool.poetry] [tool.poetry]
name = "swarms" name = "swarms"
version = "5.8.6" version = "5.8.7"
description = "Swarms - Pytorch" description = "Swarms - Pytorch"
license = "MIT" license = "MIT"
authors = ["Kye Gomez <kye@apac.ai>"] authors = ["Kye Gomez <kye@apac.ai>"]
@ -84,7 +84,7 @@ swarms = "swarms.cli.main:main"
[tool.poetry.group.lint.dependencies] [tool.poetry.group.lint.dependencies]
black = ">=23.1,<25.0" black = ">=23.1,<25.0"
ruff = ">=0.5.1,<0.6.8" ruff = ">=0.5.1,<0.6.10"
types-toml = "^0.10.8.1" types-toml = "^0.10.8.1"
types-pytz = ">=2023.3,<2025.0" types-pytz = ">=2023.3,<2025.0"
types-chardet = "^5.0.4.6" types-chardet = "^5.0.4.6"

28
swarms/prompts/prompt.py
Unescape View File

@ -13,7 +13,8 @@ from pydantic.v1 import validator
from swarms_cloud.utils.log_to_swarms_database import log_agent_data from swarms_cloud.utils.log_to_swarms_database import log_agent_data
from swarms_cloud.utils.capture_system_data import capture_system_data from swarms_cloud.utils.capture_system_data import capture_system_data
from swarms.tools.base_tool import BaseTool from swarms.tools.base_tool import BaseTool
# from swarms.agents.ape_agent import auto_generate_prompt
from typing import Any
class Prompt(BaseModel): class Prompt(BaseModel):
""" """
@ -70,11 +71,15 @@ class Prompt(BaseModel):
default="prompts", default="prompts",
description="The folder path within WORKSPACE_DIR where the prompt will be autosaved", description="The folder path within WORKSPACE_DIR where the prompt will be autosaved",
) )
auto_generate_prompt: bool = Field(
default=False,
description="Flag to enable or disable auto-generating the prompt",
)
parent_folder: str = Field( parent_folder: str = Field(
default=os.getenv("WORKSPACE_DIR"), default=os.getenv("WORKSPACE_DIR"),
description="The folder where the autosave folder is in", description="The folder where the autosave folder is in",
) )
# tools: List[callable] = None llm: Any = None
@validator("edit_history", pre=True, always=True) @validator("edit_history", pre=True, always=True)
def initialize_history(cls, v, values): def initialize_history(cls, v, values):
@ -86,6 +91,15 @@ class Prompt(BaseModel):
values["content"] values["content"]
] # Store initial version in history ] # Store initial version in history
return v return v
def __init__(self, **data):
super().__init__(**data)
if self.autosave:
self._autosave()
if self.auto_generate_prompt and self.llm:
self.auto_generate_prompt()
def edit_prompt(self, new_content: str) -> None: def edit_prompt(self, new_content: str) -> None:
""" """
@ -238,6 +252,16 @@ class Prompt(BaseModel):
with open(file_path, "w") as file: with open(file_path, "w") as file:
json.dump(self.model_dump(), file) json.dump(self.model_dump(), file)
logger.info(f"Autosaved prompt {self.id} to {file_path}.") logger.info(f"Autosaved prompt {self.id} to {file_path}.")
# def auto_generate_prompt(self):
# logger.info(f"Auto-generating prompt for {self.name}")
# task = self.name + " " + self.description + " " + self.content
# prompt = auto_generate_prompt(task, llm=self.llm, max_tokens=4000, use_second_sys_prompt=True)
# logger.info("Generated prompt successfully, updating content")
# self.edit_prompt(prompt)
# logger.info("Prompt content updated")
# return "Prompt auto-generated successfully."
class Config: class Config:
"""Pydantic configuration for better JSON serialization.""" """Pydantic configuration for better JSON serialization."""

370
swarms/structs/agent_rag.py
Unescape View File

@ -1,4 +1,3 @@
import os
from typing import List, Optional from typing import List, Optional
import chromadb import chromadb
@ -8,8 +7,17 @@ from typing import Union, Callable, Any
from swarms import Agent from swarms import Agent
class AgentRAG: class AgentRouter:
"""A vector database for storing and retrieving agents based on their characteristics.""" """
Initialize the AgentRouter.
Args:
collection_name (str): Name of the collection in the vector database.
persist_directory (str): Directory to persist the vector database.
n_agents (int): Number of agents to return in queries.
*args: Additional arguments to pass to the chromadb Client.
**kwargs: Additional keyword arguments to pass to the chromadb Client.
"""
def __init__( def __init__(
self, self,
@ -19,12 +27,6 @@ class AgentRAG:
*args, *args,
**kwargs, **kwargs,
): ):
"""
Initialize the AgentRAG.
Args:
persist_directory (str): The directory to persist the ChromaDB data.
"""
self.collection_name = collection_name self.collection_name = collection_name
self.n_agents = n_agents self.n_agents = n_agents
self.persist_directory = persist_directory self.persist_directory = persist_directory
@ -44,6 +46,9 @@ class AgentRAG:
Args: Args:
agent (Agent): The agent to add. agent (Agent): The agent to add.
Raises:
Exception: If there's an error adding the agent to the vector database.
""" """
try: try:
agent_text = f"{agent.name} {agent.description} {agent.system_prompt}" agent_text = f"{agent.name} {agent.description} {agent.system_prompt}"
@ -65,6 +70,12 @@ class AgentRAG:
def add_agents( def add_agents(
self, agents: List[Union[Agent, Callable, Any]] self, agents: List[Union[Agent, Callable, Any]]
) -> None: ) -> None:
"""
Add multiple agents to the vector database.
Args:
agents (List[Union[Agent, Callable, Any]]): List of agents to add.
"""
for agent in agents: for agent in agents:
self.add_agent(agent) self.add_agent(agent)
@ -108,9 +119,14 @@ class AgentRAG:
Args: Args:
task (str): The task description. task (str): The task description.
*args: Additional arguments to pass to the collection.query method.
**kwargs: Additional keyword arguments to pass to the collection.query method.
Returns: Returns:
Optional[Agent]: The best matching agent, if found. Optional[Agent]: The best matching agent, if found.
Raises:
Exception: If there's an error finding the best agent.
""" """
try: try:
results = self.collection.query( results = self.collection.query(
@ -150,171 +166,171 @@ class AgentRAG:
raise raise
# Example usage # # Example usage
if __name__ == "__main__": # if __name__ == "__main__":
from dotenv import load_dotenv # from dotenv import load_dotenv
from swarm_models import OpenAIChat # from swarm_models import OpenAIChat
load_dotenv() # load_dotenv()
# Get the OpenAI API key from the environment variable # # Get the OpenAI API key from the environment variable
api_key = os.getenv("GROQ_API_KEY") # api_key = os.getenv("GROQ_API_KEY")
# Model # # Model
model = OpenAIChat( # model = OpenAIChat(
openai_api_base="https://api.groq.com/openai/v1", # openai_api_base="https://api.groq.com/openai/v1",
openai_api_key=api_key, # openai_api_key=api_key,
model_name="llama-3.1-70b-versatile", # model_name="llama-3.1-70b-versatile",
temperature=0.1, # temperature=0.1,
) # )
# Initialize the vector database # # Initialize the vector database
vector_db = AgentRAG() # vector_db = AgentRouter()
# Define specialized system prompts for each agent # # Define specialized system prompts for each agent
DATA_EXTRACTOR_PROMPT = """You are a highly specialized private equity agent focused on data extraction from various documents. Your expertise includes: # DATA_EXTRACTOR_PROMPT = """You are a highly specialized private equity agent focused on data extraction from various documents. Your expertise includes:
1. Extracting key financial metrics (revenue, EBITDA, growth rates, etc.) from financial statements and reports # 1. Extracting key financial metrics (revenue, EBITDA, growth rates, etc.) from financial statements and reports
2. Identifying and extracting important contract terms from legal documents # 2. Identifying and extracting important contract terms from legal documents
3. Pulling out relevant market data from industry reports and analyses # 3. Pulling out relevant market data from industry reports and analyses
4. Extracting operational KPIs from management presentations and internal reports # 4. Extracting operational KPIs from management presentations and internal reports
5. Identifying and extracting key personnel information from organizational charts and bios # 5. Identifying and extracting key personnel information from organizational charts and bios
Provide accurate, structured data extracted from various document types to support investment analysis.""" # Provide accurate, structured data extracted from various document types to support investment analysis."""
SUMMARIZER_PROMPT = """You are an expert private equity agent specializing in summarizing complex documents. Your core competencies include: # SUMMARIZER_PROMPT = """You are an expert private equity agent specializing in summarizing complex documents. Your core competencies include:
1. Distilling lengthy financial reports into concise executive summaries # 1. Distilling lengthy financial reports into concise executive summaries
2. Summarizing legal documents, highlighting key terms and potential risks # 2. Summarizing legal documents, highlighting key terms and potential risks
3. Condensing industry reports to capture essential market trends and competitive dynamics # 3. Condensing industry reports to capture essential market trends and competitive dynamics
4. Summarizing management presentations to highlight key strategic initiatives and projections # 4. Summarizing management presentations to highlight key strategic initiatives and projections
5. Creating brief overviews of technical documents, emphasizing critical points for non-technical stakeholders # 5. Creating brief overviews of technical documents, emphasizing critical points for non-technical stakeholders
Deliver clear, concise summaries that capture the essence of various documents while highlighting information crucial for investment decisions.""" # Deliver clear, concise summaries that capture the essence of various documents while highlighting information crucial for investment decisions."""
FINANCIAL_ANALYST_PROMPT = """You are a specialized private equity agent focused on financial analysis. Your key responsibilities include: # FINANCIAL_ANALYST_PROMPT = """You are a specialized private equity agent focused on financial analysis. Your key responsibilities include:
1. Analyzing historical financial statements to identify trends and potential issues # 1. Analyzing historical financial statements to identify trends and potential issues
2. Evaluating the quality of earnings and potential adjustments to EBITDA # 2. Evaluating the quality of earnings and potential adjustments to EBITDA
3. Assessing working capital requirements and cash flow dynamics # 3. Assessing working capital requirements and cash flow dynamics
4. Analyzing capital structure and debt capacity # 4. Analyzing capital structure and debt capacity
5. Evaluating financial projections and underlying assumptions # 5. Evaluating financial projections and underlying assumptions
Provide thorough, insightful financial analysis to inform investment decisions and valuation.""" # Provide thorough, insightful financial analysis to inform investment decisions and valuation."""
MARKET_ANALYST_PROMPT = """You are a highly skilled private equity agent specializing in market analysis. Your expertise covers: # MARKET_ANALYST_PROMPT = """You are a highly skilled private equity agent specializing in market analysis. Your expertise covers:
1. Analyzing industry trends, growth drivers, and potential disruptors # 1. Analyzing industry trends, growth drivers, and potential disruptors
2. Evaluating competitive landscape and market positioning # 2. Evaluating competitive landscape and market positioning
3. Assessing market size, segmentation, and growth potential # 3. Assessing market size, segmentation, and growth potential
4. Analyzing customer dynamics, including concentration and loyalty # 4. Analyzing customer dynamics, including concentration and loyalty
5. Identifying potential regulatory or macroeconomic impacts on the market # 5. Identifying potential regulatory or macroeconomic impacts on the market
Deliver comprehensive market analysis to assess the attractiveness and risks of potential investments.""" # Deliver comprehensive market analysis to assess the attractiveness and risks of potential investments."""
OPERATIONAL_ANALYST_PROMPT = """You are an expert private equity agent focused on operational analysis. Your core competencies include: # OPERATIONAL_ANALYST_PROMPT = """You are an expert private equity agent focused on operational analysis. Your core competencies include:
1. Evaluating operational efficiency and identifying improvement opportunities # 1. Evaluating operational efficiency and identifying improvement opportunities
2. Analyzing supply chain and procurement processes # 2. Analyzing supply chain and procurement processes
3. Assessing sales and marketing effectiveness # 3. Assessing sales and marketing effectiveness
4. Evaluating IT systems and digital capabilities # 4. Evaluating IT systems and digital capabilities
5. Identifying potential synergies in merger or add-on acquisition scenarios # 5. Identifying potential synergies in merger or add-on acquisition scenarios
Provide detailed operational analysis to uncover value creation opportunities and potential risks.""" # Provide detailed operational analysis to uncover value creation opportunities and potential risks."""
# Initialize specialized agents # # Initialize specialized agents
data_extractor_agent = Agent( # data_extractor_agent = Agent(
agent_name="Data-Extractor", # agent_name="Data-Extractor",
system_prompt=DATA_EXTRACTOR_PROMPT, # system_prompt=DATA_EXTRACTOR_PROMPT,
llm=model, # llm=model,
max_loops=1, # max_loops=1,
autosave=True, # autosave=True,
verbose=True, # verbose=True,
dynamic_temperature_enabled=True, # dynamic_temperature_enabled=True,
saved_state_path="data_extractor_agent.json", # saved_state_path="data_extractor_agent.json",
user_name="pe_firm", # user_name="pe_firm",
retry_attempts=1, # retry_attempts=1,
context_length=200000, # context_length=200000,
output_type="string", # output_type="string",
) # )
summarizer_agent = Agent( # summarizer_agent = Agent(
agent_name="Document-Summarizer", # agent_name="Document-Summarizer",
system_prompt=SUMMARIZER_PROMPT, # system_prompt=SUMMARIZER_PROMPT,
llm=model, # llm=model,
max_loops=1, # max_loops=1,
autosave=True, # autosave=True,
verbose=True, # verbose=True,
dynamic_temperature_enabled=True, # dynamic_temperature_enabled=True,
saved_state_path="summarizer_agent.json", # saved_state_path="summarizer_agent.json",
user_name="pe_firm", # user_name="pe_firm",
retry_attempts=1, # retry_attempts=1,
context_length=200000, # context_length=200000,
output_type="string", # output_type="string",
) # )
financial_analyst_agent = Agent( # financial_analyst_agent = Agent(
agent_name="Financial-Analyst", # agent_name="Financial-Analyst",
system_prompt=FINANCIAL_ANALYST_PROMPT, # system_prompt=FINANCIAL_ANALYST_PROMPT,
llm=model, # llm=model,
max_loops=1, # max_loops=1,
autosave=True, # autosave=True,
verbose=True, # verbose=True,
dynamic_temperature_enabled=True, # dynamic_temperature_enabled=True,
saved_state_path="financial_analyst_agent.json", # saved_state_path="financial_analyst_agent.json",
user_name="pe_firm", # user_name="pe_firm",
retry_attempts=1, # retry_attempts=1,
context_length=200000, # context_length=200000,
output_type="string", # output_type="string",
) # )
market_analyst_agent = Agent( # market_analyst_agent = Agent(
agent_name="Market-Analyst", # agent_name="Market-Analyst",
system_prompt=MARKET_ANALYST_PROMPT, # system_prompt=MARKET_ANALYST_PROMPT,
llm=model, # llm=model,
max_loops=1, # max_loops=1,
autosave=True, # autosave=True,
verbose=True, # verbose=True,
dynamic_temperature_enabled=True, # dynamic_temperature_enabled=True,
saved_state_path="market_analyst_agent.json", # saved_state_path="market_analyst_agent.json",
user_name="pe_firm", # user_name="pe_firm",
retry_attempts=1, # retry_attempts=1,
context_length=200000, # context_length=200000,
output_type="string", # output_type="string",
) # )
operational_analyst_agent = Agent( # operational_analyst_agent = Agent(
agent_name="Operational-Analyst", # agent_name="Operational-Analyst",
system_prompt=OPERATIONAL_ANALYST_PROMPT, # system_prompt=OPERATIONAL_ANALYST_PROMPT,
llm=model, # llm=model,
max_loops=1, # max_loops=1,
autosave=True, # autosave=True,
verbose=True, # verbose=True,
dynamic_temperature_enabled=True, # dynamic_temperature_enabled=True,
saved_state_path="operational_analyst_agent.json", # saved_state_path="operational_analyst_agent.json",
user_name="pe_firm", # user_name="pe_firm",
retry_attempts=1, # retry_attempts=1,
context_length=200000, # context_length=200000,
output_type="string", # output_type="string",
) # )
# Create agents (using the agents from the original code) # # Create agents (using the agents from the original code)
agents_to_add = [ # agents_to_add = [
data_extractor_agent, # data_extractor_agent,
summarizer_agent, # summarizer_agent,
financial_analyst_agent, # financial_analyst_agent,
market_analyst_agent, # market_analyst_agent,
operational_analyst_agent, # operational_analyst_agent,
] # ]
# Add agents to the vector database # # Add agents to the vector database
for agent in agents_to_add: # for agent in agents_to_add:
vector_db.add_agent(agent) # vector_db.add_agent(agent)
# Example task # # Example task
task = "Analyze the financial statements of a potential acquisition target and identify key growth drivers." # task = "Analyze the financial statements of a potential acquisition target and identify key growth drivers."
# Find the best agent for the task # # Find the best agent for the task
best_agent = vector_db.find_best_agent(task) # best_agent = vector_db.find_best_agent(task)
if best_agent: # if best_agent:
logger.info(f"Best agent for the task: {best_agent.name}") # logger.info(f"Best agent for the task: {best_agent.name}")
# Use the best agent to perform the task # # Use the best agent to perform the task
result = best_agent.run(task) # result = best_agent.run(task)
print(f"Task result: {result}") # print(f"Task result: {result}")
# Update the agent's history in the database # # Update the agent's history in the database
vector_db.update_agent_history(best_agent.name) # vector_db.update_agent_history(best_agent.name)
else: # else:
print("No suitable agent found for the task.") # print("No suitable agent found for the task.")
# Save the vector database # # Save the vector database

20
swarms/structs/auto_swarm_builder.py
Unescape View File

@ -0,0 +1,20 @@
from typing import List, Any, Dict, Optional
class AutoSwarmBuilder:
def __init__(self, task: str, num_agents: int = 10, batch_size: int = 10):
self.task = task
self.num_agents = num_agents
self.batch_size = batch_size
def run(self, task: str, image_url: str = None, *args, **kwargs):
pass
def _create_swarm(self):
pass
def _create_agents(self):
pass
def _run_agents(self):
pass

3
swarms/structs/concurrent_workflow.py
Unescape View File

@ -138,7 +138,6 @@ class ConcurrentWorkflow(BaseSwarm):
self.reliability_check() self.reliability_check()
def reliability_check(self): def reliability_check(self):
try: try:
logger.info("Starting reliability checks") logger.info("Starting reliability checks")
@ -395,7 +394,7 @@ class ConcurrentWorkflow(BaseSwarm):
""" """
if task is not None: if task is not None:
self.tasks.append(task) self.tasks.append(task)
try: try:
logger.info(f"Attempting to run on device: {device}") logger.info(f"Attempting to run on device: {device}")
if device == "cpu": if device == "cpu":

71
swarms/structs/run_agents_in_parallel_async_multiprocess.py
Unescape View File

@ -1,74 +1,81 @@
import os
import asyncio import asyncio
from typing import List, Any
from swarms import Agent from swarms import Agent
from swarm_models import OpenAIChat
import uvloop
from multiprocessing import cpu_count from multiprocessing import cpu_count
from swarms.utils.calculate_func_metrics import profile_func from swarms.utils.calculate_func_metrics import profile_func
from typing import List
# Use uvloop for faster asyncio event loop # Use uvloop for faster asyncio event loop
asyncio.set_event_loop_policy(uvloop.EventLoopPolicy()) asyncio.set_event_loop_policy(uvloop.EventLoopPolicy())
# Get the OpenAI API key from the environment variable
api_key = os.getenv("OPENAI_API_KEY")
# Get the OpenAI API key from the environment variable def run_single_agent(agent: Agent, task: str) -> Any:
api_key = os.getenv("OPENAI_API_KEY") """
Run a single agent on the given task.
# Create an instance of the OpenAIChat class (can be reused)
model = OpenAIChat(
api_key=api_key, model_name="gpt-4o-mini", temperature=0.1
)
Args:
agent (Agent): The agent to run.
task (str): The task for the agent to perform.
# Function to run a single agent on the task (synchronous) Returns:
def run_single_agent(agent, task): Any: The result of the agent's execution.
"""
return agent.run(task) return agent.run(task)
# Asynchronous wrapper for agent tasks async def run_agent_async(agent: Agent, task: str) -> Any:
async def run_agent_async(agent, task): """
Asynchronous wrapper for agent tasks.
Args:
agent (Agent): The agent to run asynchronously.
task (str): The task for the agent to perform.
Returns:
Any: The result of the agent's execution.
"""
loop = asyncio.get_event_loop() loop = asyncio.get_event_loop()
return await loop.run_in_executor( return await loop.run_in_executor(
None, run_single_agent, agent, task None, run_single_agent, agent, task
) )
# Asynchronous function to run agents concurrently async def run_agents_concurrently_async(
async def run_agents_concurrently_async(agents, task: str): agents: List[Agent], task: str
) -> List[Any]:
""" """
Run multiple agents concurrently on the same task with optimized performance. Run multiple agents concurrently on the same task with optimized performance.
:param agents: List of Agent instances to run concurrently. Args:
:param task: The task string to execute by all agents. agents (List[Agent]): List of Agent instances to run concurrently.
:return: A list of outputs from each agent. task (str): The task string to execute by all agents.
"""
# Run all agents asynchronously using asyncio.gather Returns:
List[Any]: A list of outputs from each agent.
"""
results = await asyncio.gather( results = await asyncio.gather(
*(run_agent_async(agent, task) for agent in agents) *(run_agent_async(agent, task) for agent in agents)
) )
return results return results
# Function to manage the overall process and batching
@profile_func @profile_func
def run_agents_concurrently_multiprocess( def run_agents_concurrently_multiprocess(
agents: List[Agent], task: str, batch_size: int = cpu_count() agents: List[Agent], task: str, batch_size: int = cpu_count()
): ) -> List[Any]:
""" """
Manage and run multiple agents concurrently in batches, with optimized performance. Manage and run multiple agents concurrently in batches, with optimized performance.
:param agents: List of Agent instances to run concurrently. Args:
:param task: The task string to execute by all agents. agents (List[Agent]): List of Agent instances to run concurrently.
:param batch_size: Number of agents to run in parallel in each batch. task (str): The task string to execute by all agents.
:return: A list of outputs from each agent. batch_size (int, optional): Number of agents to run in parallel in each batch.
""" Defaults to the number of CPU cores.
Returns:
List[Any]: A list of outputs from each agent.
"""
results = [] results = []
loop = asyncio.get_event_loop() loop = asyncio.get_event_loop()
# batch_size = cpu_count()
# Process agents in batches to avoid overwhelming system resources # Process agents in batches to avoid overwhelming system resources
for i in range(0, len(agents), batch_size): for i in range(0, len(agents), batch_size):

Write Preview
Loading…
Cancel
Save