What Is DALL-E? OpenAI's Image Generator, Explained

DALL-E is the text-to-image AI built by OpenAI, the company behind ChatGPT. You describe a picture in plain language — a prompt — and DALL-E generates an original image to match. Its biggest convenience is that it’s built right into ChatGPT, so you can ask for an image in the same chat where you write and brainstorm. If the underlying technology is new to you, our explainer on how AI chatbots work covers the same kind of AI applied to text.

How DALL-E works

You write a prompt — “a cozy bookstore café in autumn, soft lighting” — and DALL-E produces an image from it. It’s trained on huge numbers of image-and-caption pairs, so it has learned how words correspond to visual concepts. Because it lives inside ChatGPT, you can also refine results conversationally: ask for changes in tone, add or remove elements, or adjust the style in follow-up messages.

What makes DALL-E different

• Built into ChatGPT. No separate app or workflow — you generate images in the same place you already chat, which makes it the easiest on-ramp for newcomers.
• Strong prompt understanding. It’s good at following detailed, natural-language descriptions, including text and specific arrangements of objects.
• Conversational editing. You can iterate by chatting rather than learning special commands.

DALL-E vs Midjourney

The two best-known image generators have different strengths. DALL-E is the most convenient and beginner-friendly thanks to its ChatGPT integration and natural prompt handling. Midjourney is often praised for more striking, stylized and cinematic results, but lives in its own app — see what Midjourney is. Many creators use both depending on the job.

Is DALL-E free?

DALL-E image generation is generally available through ChatGPT’s paid plans, with limits on how many images you can create in a period. Some access may appear in OpenAI’s and Microsoft’s other products too. Availability and limits change often, so check OpenAI for current details.

Choosing an AI image tool

DALL-E is one of several strong options, and the best choice depends on the look you want, your budget and whether you need commercial rights. For a side-by-side look at the leading tools, see our roundup of the best AI image generators.

FAQ

Who makes DALL-E?

DALL-E is made by OpenAI, the same company behind ChatGPT. It’s OpenAI’s text-to-image model and is built into ChatGPT.

How does DALL-E work?

You write a text prompt describing the image you want, and DALL-E generates it using a model trained on huge numbers of image-and-caption pairs. Inside ChatGPT, you can refine the result by chatting.

Is DALL-E free?

Image generation is generally part of ChatGPT’s paid plans, with limits on how many you can create. Availability can vary across OpenAI’s and Microsoft’s products, so check OpenAI for current details.

DALL-E or Midjourney — which is better?

Neither is universally better. DALL-E is the most convenient and beginner-friendly via ChatGPT; Midjourney is prized for more stylized, cinematic images. Many people use both. See our best AI image generators comparison to choose.

---

The AI field moves fast and tool capabilities change frequently. This explainer reflects the current landscape and is reviewed periodically.