How to beginner · 3 min read

How to edit image with OpenAI DALL-E

Q: How to edit image with OpenAI DALL-E

Use the OpenAI Python SDK's client.images.edits.create method with your original image, an optional mask image, and a prompt describing the edit. This sends a request to the DALL-E image editing endpoint to generate the edited image.

Quick answer

Use the OpenAI Python SDK's client.images.edits.create method with your original image, an optional mask image, and a prompt describing the edit. This sends a request to the DALL-E image editing endpoint to generate the edited image.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0
An original image file and optional mask image file

Setup

Install the OpenAI Python SDK and set your API key as an environment variable.

Run pip install openai to install the SDK.
Set your API key in your shell: export OPENAI_API_KEY='your_api_key_here' (Linux/macOS) or setx OPENAI_API_KEY "your_api_key_here" (Windows).

bash

pip install openai

Step by step

This example edits an image by providing the original image, a mask image to specify the editable area, and a prompt describing the desired change.

python

import os
from openai import OpenAI

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

# Paths to your images
original_image_path = "./original.png"
mask_image_path = "./mask.png"

response = client.images.edits.create(
    model="dall-e-3",
    image=open(original_image_path, "rb"),
    mask=open(mask_image_path, "rb"),
    prompt="Add a red hat to the person",
    n=1,
    size="1024x1024"
)

edited_image_url = response.data[0].url
print("Edited image URL:", edited_image_url)

output

Edited image URL: https://openai-generated-image-url.com/edited-image.png

Common variations

Change model to other DALL-E versions if available.
Adjust size to "256x256" or "512x512" for smaller images.
Use asynchronous calls with asyncio for non-blocking requests.
Generate multiple edits by increasing n.

python

import os
import asyncio
from openai import OpenAI

async def edit_image_async():
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    response = await client.images.edits.acreate(
        model="dall-e-3",
        image=open("./original.png", "rb"),
        mask=open("./mask.png", "rb"),
        prompt="Add sunglasses",
        n=2,
        size="512x512"
    )
    for i, data in enumerate(response.data):
        print(f"Edited image {i+1} URL:", data.url)

asyncio.run(edit_image_async())

output

Edited image 1 URL: https://openai-generated-image-url.com/edited1.png
Edited image 2 URL: https://openai-generated-image-url.com/edited2.png

Troubleshooting

If you get a FileNotFoundError, verify your image paths are correct.
If the API returns an authentication error, check your OPENAI_API_KEY environment variable.
For InvalidRequestError, ensure your images meet size and format requirements (PNG, JPG).
Network errors may require retry logic or checking your internet connection.

✅

Key Takeaways

Use client.images.edits.create with original image, mask, and prompt to edit images with DALL-E.
Always provide a mask image to specify editable regions for precise edits.
Use asynchronous calls for efficient image editing in concurrent applications.
Check environment variables and file paths to avoid common errors.
Adjust model, size, and n parameters to customize output.

Verified 2026-04 · dall-e-3

Verify ↗