How to beginner · 3 min read

How to specify output format in prompts

Quick answer

To specify output format in prompts, explicitly instruct the AI to respond in a desired structure such as JSON, XML, or markdown by including format instructions in the prompt text. Use clear examples or templates within the prompt to guide the model's output format precisely.

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 for secure access.

bash

pip install openai>=1.0

Step by step

Use explicit instructions in your prompt to specify the output format. For example, ask the model to return a JSON object with specific keys. This ensures the response is structured and easy to parse.

python

import os
from openai import OpenAI

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

prompt = '''
Generate a JSON object with the following keys: "name", "age", and "city".
Example:
{
  "name": "Alice",
  "age": 30,
  "city": "New York"
}
'''

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": prompt}]
)

print(response.choices[0].message.content)

output

{
  "name": "Alice",
  "age": 30,
  "city": "New York"
}

Common variations

You can specify other formats like XML or markdown by changing the instructions. For streaming or async calls, adapt the SDK usage accordingly. Different models may interpret format instructions with varying strictness.

python

import os
from anthropic import Anthropic

client = Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])

system_prompt = "You are a helpful assistant."
user_prompt = '''
Please provide the output in XML format with tags: &lt;name&gt;, &lt;age&gt;, &lt;city&gt;.
Example:
<person>
  <name>Alice</name>
  <age>30</age>
  <city>New York</city>
</person>
'''

message = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    max_tokens=500,
    system=system_prompt,
    messages=[{"role": "user", "content": user_prompt}]
)

print(message.content[0].text)

output

<person>
  <name>Alice</name>
  <age>30</age>
  <city>New York</city>
</person>

Troubleshooting

If the output format is inconsistent, try adding explicit examples or use delimiters like triple backticks to enforce code blocks. Also, verify the model supports your requested format and check for prompt length limits.

✅

Key Takeaways

Always include explicit instructions and examples in your prompt to specify output format.
Use code block delimiters like triple backticks to help the model produce clean, parseable output.
Different models and SDKs support format instructions differently; test and adjust accordingly.

Verified 2026-04 · gpt-4o, claude-3-5-sonnet-20241022

Verify ↗