Latest models
- DALL-E 2 and DALL-E 3 (May 12, 2026) Retired
Removed from the OpenAI API on May 12, 2026. OpenAI recommends gpt-image-2, gpt-image-1 or gpt-image-1-mini instead. Source: OpenAI deprecations ↗
Checked October 7, 2026. See the full AI model tracker.
What we like
- The prompting style it popularised still works: describe the scene in plain sentences
- Its replacements are available in ChatGPT and the API
Watch out for
- The DALL-E 2 and 3 API models are no longer available
- Old tutorials that name dall-e-3 need updating
DALL-E was OpenAI’s text-to-image model, and many of the images on this site were made with it. It has been replaced.
What happened
OpenAI’s deprecations page lists DALL-E 2 and DALL-E 3 as removed from the API on May 12, 2026. OpenAI recommends gpt-image-2, gpt-image-1 or gpt-image-1-mini instead. OpenAI’s current image models are covered on our GPT Image page.
What still works
The prompts below were written for DALL-E and work as plain-language prompts in OpenAI’s newer image models. Describe the subject, setting, light and style in full sentences, and name any text you want in the image.
If you followed an old tutorial
Replace dall-e-3 with a GPT Image model name in code, and expect different default behaviour. Our older DALL-E posts, such as how to generate realistic images with DALL-E, describe the 2024 interface.