How to beginner · 3 min read

How to limit agent iterations in python

Quick answer

To limit agent iterations in Python, set a maximum iteration count in your agent's control loop or use built-in parameters like max_iterations if supported by your agent framework. This prevents infinite loops and controls resource usage effectively.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the openai Python package and set your API key as an environment variable to authenticate requests.

bash

pip install openai

Step by step

Use a loop with a counter to limit the number of iterations your agent runs. Below is a simple example using OpenAI's gpt-4o model to simulate an agent that stops after 3 iterations.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

max_iterations = 3
iteration = 0

while iteration < max_iterations:
    response = client.chat.completions.create(
        model="gpt-4o",
        messages=[{"role": "user", "content": f"Iteration {iteration + 1}: Say hello!"}]
    )
    print(response.choices[0].message.content)
    iteration += 1

output

Hello!
Hello!
Hello!

Common variations

You can limit iterations asynchronously or use agent frameworks like LangChain that support max_iterations parameters directly. For example, LangChain agents accept max_iterations to stop automatically.

python

from langchain_openai import ChatOpenAI
from langchain.agents import initialize_agent, Tool

llm = ChatOpenAI(model_name="gpt-4o", temperature=0)

tools = [Tool(name="Echo", func=lambda x: x, description="Echo input")]

agent = initialize_agent(tools, llm, max_iterations=2, agent_type="zero-shot-react-description")

result = agent.run("Say hello twice")
print(result)

output

Hello!
Hello!

Troubleshooting

If your agent runs indefinitely, ensure your iteration limit variable is correctly incremented and checked. In frameworks, verify you pass max_iterations correctly and update to the latest SDK version.

✅

Key Takeaways

Use a loop counter or max_iterations parameter to limit agent runs.
LangChain and similar frameworks support built-in iteration limits for agents.
Always increment and check your iteration variable to avoid infinite loops.

Verified 2026-04 · gpt-4o, langchain_openai ChatOpenAI

Verify ↗