How to beginner · 3 min read

How to use async execution in CrewAI

Quick answer

Use CrewAI's async client methods by importing the async API client and leveraging Python's asyncio to run AI calls concurrently. This enables non-blocking execution of multiple AI tasks, improving throughput and responsiveness.

PREREQUISITES

Python 3.8+
CrewAI SDK installed (pip install crewai)
API key for CrewAI set in environment variable CREWAI_API_KEY

Setup

Install the CrewAI SDK and set your API key as an environment variable to enable authentication.

bash

pip install crewai

Step by step

Use the async client from CrewAI with asyncio to run multiple AI completions concurrently. Below is a complete example demonstrating async execution.

python

import os
import asyncio
from crewai import AsyncCrewAIClient

async def main():
    client = AsyncCrewAIClient(api_key=os.environ["CREWAI_API_KEY"])

    async def get_completion(prompt):
        response = await client.chat.completions.create(
            model="gpt-4o",
            messages=[{"role": "user", "content": prompt}]
        )
        return response.choices[0].message.content

    prompts = [
        "Explain async programming in Python.",
        "What is CrewAI?",
        "Benefits of async execution in AI calls."
    ]

    # Run all completions concurrently
    results = await asyncio.gather(*(get_completion(p) for p in prompts))

    for i, result in enumerate(results, 1):
        print(f"Response {i}: {result}\n")

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

output

Response 1: Async programming in Python allows you to write concurrent code using async and await keywords, enabling non-blocking operations.

Response 2: CrewAI is an AI platform that provides easy-to-use APIs for integrating AI models with support for async execution.

Response 3: Async execution improves performance by running multiple AI calls concurrently, reducing wait times and increasing throughput.

Common variations

You can also use synchronous calls if async is not needed, or switch models by changing the model parameter. CrewAI supports streaming responses and other advanced features with async methods.

python

import os
from crewai import CrewAIClient

client = CrewAIClient(api_key=os.environ["CREWAI_API_KEY"])
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Hello from sync call"}]
)
print(response.choices[0].message.content)

output

Hello from sync call

Troubleshooting

If you get authentication errors, verify your CREWAI_API_KEY environment variable is set correctly.
For asyncio errors, ensure you are running Python 3.8 or higher and using asyncio.run() properly.
If responses are slow, check your network connection and CrewAI service status.

✅

Key Takeaways

Use CrewAI's async client with Python's asyncio for concurrent AI calls.
Set your API key in the environment variable CREWAI_API_KEY for authentication.
Async execution reduces latency by running multiple requests simultaneously.

Verified 2026-04 · gpt-4o, gpt-4o-mini

Verify ↗