Code beginner · 3 min read

How to install CrewAI in python

Q: How to install CrewAI in python

Install CrewAI in Python using pip install crewai and import it with import crewai to start using its API.

Direct answer

Install CrewAI in Python using pip install crewai and import it with import crewai to start using its API.

Setup

Install

bash

pip install crewai

Env vars

CREWAI_API_KEY

Imports

python

import os
import crewai

Examples

inInitialize CrewAI client and get a simple completion for 'Hello, world!'

outResponse: Hello, world! How can I assist you today?

inUse CrewAI to generate a short summary of a text about AI advancements.

outResponse: AI advancements have accelerated rapidly, enabling smarter applications across industries.

inCall CrewAI with an empty prompt to test error handling.

outError: Prompt cannot be empty. Please provide valid input.

Integration steps

Install the CrewAI Python package using pip.
Set your API key in the environment variable CREWAI_API_KEY.
Import the crewai module in your Python script.
Initialize the CrewAI client with the API key from os.environ.
Call the appropriate CrewAI method with your input prompt.
Extract and use the response text from the API call.

Full code

python

import os
import crewai

# Initialize CrewAI client with API key from environment
client = crewai.Client(api_key=os.environ['CREWAI_API_KEY'])

# Define a prompt
prompt = "Hello, CrewAI!"

# Call the completion method
response = client.messages.create(prompt=prompt, max_tokens=50)

# Print the response text
print("Response:", response.text)

output

Response: Hello, CrewAI! How can I assist you today?

API trace

Request

json

{"prompt": "Hello, CrewAI!", "max_tokens": 50}

Response

json

{"text": "Hello, CrewAI! How can I assist you today?", "usage": {"tokens": 15}}

Extractresponse.text

Variants

Streaming response version ›

Use streaming to display partial results immediately for long responses or better user experience.

python

import os
import crewai

client = crewai.Client(api_key=os.environ['CREWAI_API_KEY'])
prompt = "Tell me a story about AI."

for chunk in client.completions.stream(prompt=prompt, max_tokens=100):
    print(chunk.text, end='', flush=True)
print()

Async version ›

Use async calls to handle multiple concurrent requests efficiently in an async Python environment.

python

import os
import asyncio
import crewai

async def main():
    client = crewai.Client(api_key=os.environ['CREWAI_API_KEY'])
    prompt = "Explain quantum computing in simple terms."
    response = await client.completions.acreate(prompt=prompt, max_tokens=60)
    print("Response:", response.text)

asyncio.run(main())

Alternative model usage ›

Use a different model variant for improved accuracy or specialized tasks.

python

import os
import crewai

client = crewai.Client(api_key=os.environ['CREWAI_API_KEY'])
prompt = "Summarize the latest AI trends."
response = client.messages.create(prompt=prompt, max_tokens=50, model="crewai-advanced-v2")
print("Response:", response.text)

Performance

Latency~500ms for typical completion calls

Cost~$0.0015 per 100 tokens generated

Rate limitsDefault tier: 300 requests per minute, 50,000 tokens per minute

Use concise prompts to reduce token usage.
Limit <code>max_tokens</code> to the minimum needed for your task.
Reuse context when possible to avoid repeated tokens.

Approach	Latency	Cost/call	Best for
Standard completion	~500ms	~$0.0015	General purpose text generation
Streaming completion	~300ms initial + streaming	~$0.0015	Long outputs with better UX
Async completion	~500ms	~$0.0015	Concurrent requests in async apps

✓

Quick tip

Always set your CrewAI API key in the environment variable <code>CREWAI_API_KEY</code> to keep credentials secure.

⚠

Common mistake

Beginners often forget to set the <code>CREWAI_API_KEY</code> environment variable, causing authentication errors.

Verified 2026-04 · crewai-advanced-v2

Verify ↗