How to beginner · 3 min read

How to translate product listings with AI

Q: How to translate product listings with AI

Use a large language model like gpt-4o to translate product listings by sending the original text as input and requesting the target language output. Automate this with the OpenAI Python SDK to batch process listings for multilingual ecommerce sites.

Quick answer

Use a large language model like gpt-4o to translate product listings by sending the original text as input and requesting the target language output. Automate this with the OpenAI Python SDK to batch process listings for multilingual ecommerce sites.

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

output

Collecting openai
  Downloading openai-1.x.x-py3-none-any.whl
Installing collected packages: openai
Successfully installed openai-1.x.x

Step by step

This example shows how to translate a product listing from English to Spanish using gpt-4o. It sends the original product description as a prompt and requests the translation.

python

import os
from openai import OpenAI

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

product_listing = "This stylish leather wallet features multiple card slots and a secure zipper compartment."

messages = [
    {"role": "user", "content": f"Translate the following product listing to Spanish:\n{product_listing}"}
]

response = client.chat.completions.create(
    model="gpt-4o",
    messages=messages
)

translation = response.choices[0].message.content
print("Translated product listing:", translation)

output

Translated product listing: Esta elegante cartera de cuero cuenta con múltiples ranuras para tarjetas y un compartimento seguro con cremallera.

Common variations

You can translate listings to other languages by changing the prompt language. For batch processing, loop over multiple listings. Use gpt-4o-mini for cost-effective translations or claude-3-5-sonnet-20241022 from Anthropic as an alternative.

python

import os
from openai import OpenAI

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

listings = [
    "Elegant stainless steel watch with water resistance up to 50 meters.",
    "Comfortable cotton t-shirt available in multiple colors."
]

target_language = "French"

for listing in listings:
    messages = [
        {"role": "user", "content": f"Translate this product listing to {target_language}:\n{listing}"}
    ]
    response = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=messages
    )
    print(f"Original: {listing}")
    print(f"Translated: {response.choices[0].message.content}\n")

output

Original: Elegant stainless steel watch with water resistance up to 50 meters.
Translated: Montre élégante en acier inoxydable avec une résistance à l'eau jusqu'à 50 mètres.

Original: Comfortable cotton t-shirt available in multiple colors.
Translated: T-shirt en coton confortable disponible en plusieurs couleurs.

Troubleshooting

If translations are incomplete or cut off, increase max_tokens in the API call.
If you get rate limit errors, implement exponential backoff retries.
For inconsistent translations, add system instructions to specify style or tone.

✅

Key Takeaways

Use gpt-4o or gpt-4o-mini for accurate and cost-effective product listing translations.
Batch translate multiple listings by looping API calls with dynamic prompts.
Set max_tokens appropriately to avoid truncated translations.
Add system-level instructions to control translation style and tone.
Handle API rate limits with retry logic for robust production use.

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

Verify ↗