[Python] How to run multiple browser-use agents in parallel?

Currently I have this code

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
:brief: Run multiple browser-use agents in parallel
"""
import asyncio
import os

from browser_use import Agent, Browser, BrowserConfig, Controller
from browser_use.browser.context import BrowserContext, BrowserContextConfig
from langchain_openai import ChatOpenAI


async def main():
controller = Controller()
llm = ChatOpenAI(model="gpt-4o", api_key=os.getenv("OPENAI_API_KEY"))
browser_config = BrowserConfig(
chrome_instance_path="/path/to/chrome",
headless=False,
disable_security=True,
extra_chromium_args=[
"--no-first-run",
"--no-default-browser-check",
"--disable-extensions",
"--new-window",
],
)
browser = Browser(config=browser_config)
context1 = BrowserContext(browser=browser)
context2 = BrowserContext(browser=browser)

agent = Agent(
task="Go to amazon.com and search for toys without logging in",
llm=llm,
controller=controller,
browser=browser,
browser_context=context1,
)

agent2 = Agent(
task="Go to Google.com and find the latest news",
llm=llm,
controller=controller,
browser=browser,
browser_context=context2,
)

await asyncio.gather(agent.run(), agent2.run())

if __name__ == "__main__":
asyncio.run(main())


But, it does not truly run in parellel. I was considering using ProcessPoolExecutor but not sure how to mix concurrent.futures and asyncio cleanly. Essentially, the approach should be to create 2 different browser instances so that the agents can work in tandem.

Continue reading...