What is DALL-E 3?
DALL-E 3 is OpenAI’s text-to-image model for ChatGPT users and developers who want to turn written ideas into images. Compared with DALL-E 2, DALL-E 3 is designed to follow more of the detail in a description, including when the two models receive the same prompt. The page introducing DALL-E 3 points readers to ChatGPT Images 2.5 for the current ChatGPT image experience.
What DALL-E 3 can do
DALL-E 3 generates images from descriptions that range from a simple sentence to a detailed paragraph. The ChatGPT integration adds a way to develop a prompt before generating an image and to request changes afterward. Users can therefore describe what they want to see without having to write the full generation prompt themselves.
- DALL-E 3 turns written descriptions into images.
- DALL-E 3 follows prompt details more closely than DALL-E 2.
- DALL-E 3 uses detailed prompts that ChatGPT generates from user ideas.
- DALL-E 3 supports image revisions through short requests in ChatGPT.
How DALL-E 3 works
DALL-E 3 is built into the ChatGPT workflow described on the introduction page. A user starts by describing an image in ordinary language, and ChatGPT turns that idea into a tailored prompt for the model. If the resulting image is close but not quite right, the user can ask ChatGPT for a change in a few words rather than rewriting the entire description. The page also describes API access for developers.
Who DALL-E 3 is for
DALL-E 3 serves people who want to develop a visual idea through conversation and developers who prefer to work through an API. The ChatGPT workflow lets users brainstorm before settling on a prompt, then adjust the result after seeing it. OpenAI also says users do not need its permission to reprint, sell, or merchandise the images they create with DALL-E 3.
- Develop a visual idea from a short description.
- Refine an image with a follow-up request in ChatGPT.
- Create images to reprint, sell, or use on merchandise.
- Generate images through the developer API.
DALL-E 3 pricing and access
DALL-E 3 was described on its introduction page as available to all ChatGPT users and to developers through the OpenAI API. The page does not give prices, list plan tiers, or say whether API access is free. Its link to ChatGPT Images 2.5 identifies the current ChatGPT image experience, so the original access statement does not establish which model users will find there today.
DALL-E 3 strengths and limitations
DALL-E 3 offers closer adherence to written descriptions than DALL-E 2, and the ChatGPT workflow lets users refine a result without starting over with a new description. The introduction page also describes work on identifying AI-generated images. OpenAI presents its provenance classifier as an experimental internal tool, not as an image-checking feature available to users.
- DALL-E 3 is designed to decline requests for a living artist’s style.
- DALL-E 3 has safeguards against requests that name a public figure.
- The provenance classifier is described as an internal experiment.
DALL-E 3 FAQ
What is DALL-E 3?
DALL-E 3 is OpenAI’s text-to-image model for generating images from written descriptions. It follows more of the detail in a prompt than DALL-E 2, and its ChatGPT integration lets users develop prompts and request changes conversationally.
Is DALL-E 3 free?
DALL-E 3 does not have pricing listed on its introduction page. The page describes access for all ChatGPT users and for developers through the OpenAI API, but it does not give prices or explain how either access route is billed.
Does DALL-E 3 have an API?
DALL-E 3 is described on its introduction page as available to developers through the OpenAI API. That is separate from the ChatGPT workflow on the page, where users describe an image conversationally and can request changes with a short follow-up.
Can you sell images made with DALL-E 3?
DALL-E 3 images can be sold, reprinted, or used on merchandise without asking OpenAI for permission, according to the model’s introduction page. OpenAI says the images users create with DALL-E 3 are theirs to use.
How do you revise a DALL-E 3 image in ChatGPT?
DALL-E 3 supports image revisions through short follow-up requests in the ChatGPT workflow described on the introduction page. If a generated image is nearly right, a user can tell ChatGPT what to change in a few words.
From the community
No reviews yet
Be the first to review this tool. What you write helps the next person choose.



