Article Complete
AIM
DALL·E introduces 'Outpainting', a new feature that extends images
DALL·E introduces 'Outpainting', a new feature that extends images

DALL·E introduces 'Outpainting', a new feature that extends images

More than one million people use the AI system as a creative tool to generate original images or artwork from a natural language description.

Published Updated

Key takeaways

  • DALL·E's new feature 'Outpainting' allows users to extend images beyond their original borders.
  • The feature enables the creation of large-scale images while preserving existing visual elements.
  • Over one million users utilise DALL·E as a creative tool for generating original artwork.
  • Outpainting is accessible to all DALL·E users on desktop, expanding creative possibilities.
  • DALL·E uses a modified version of GPT-3 to generate images from natural language descriptions.
5 min

In a blog post yesterday, DALL·E introduced ‘Outpainting’, a new feature that helps users extend their creativity by using natural language description. Outpainting would allow users to continue an image beyond the original borders by adding visual elements in the same style, with the brush that would fill spaces beyond the original canvas. 

DALL·E’s existing feature ‘Inpainting’ enables changes within a generated image. However, with Outpainting, users can create large-scale images in any aspect ratio by extending the original image. The new feature takes into account the existing visual elements such as reflections, shadows, and textures to maintain the context of the images’ originality. 

Original: 'Girl with a Pearl Earring', Johannes Vermeer | Outpainting: August Kamp

More than one million people use the AI system as a creative tool to generate original images or artwork from a natural language description. Artists around the world have used the new Outpainting feature to create remarkable artwork, which in turn helps DALL·E understand their capabilities in the process. 

Outpainting is available to all DALL·E users on desktop. To further explore and discover, visit labs.openai.com or join the waitlist.

Revealed by OpenAI in a blog in 2021, DALL·E uses a version of GPT-3 modified to generate images. DALL-E and DALL-E 2 are machine learning models that are used to generate digital images from natural language descriptions. 

Membership

Stay ahead with AI insights.

Unlimited access to premium articles, exclusive analysis, and the stories shaping India's AI economy.

Subscribe now

AIM Americas

1190 Miraloma Way Suite P, Sunnyvale, CA 94085

AIM India

1st Floor, Sakti Statesman, Bellandur, Bengaluru – 560103

© 2026 AIM. All rights reserved.