OpenAI's hosted text-to-image model family for prompts and image edits
DALL-E is OpenAI's family of hosted image generation models, reached through ChatGPT, the OpenAI API and partner products rather than as a separate downloadable program. It converts a text description into an image, supports conversational refinement where you ask for changes in plain language, and offers editing features such as regenerating a selected region or using a reference image as guidance. Safety filtering and content policies apply to prompts and outputs. The differentiator is that generation sits inside a general assistant, so prompting, revision and description happen in one conversation.
Last updated: 2026-09-20. This site only provides an index; for exact features, pricing, and licensing, see the official website.
concept art, illustration drafts, marketing imagery, prompt-driven image editing
If you're comparing similar products, check the alternatives below, or browse all tools in the AI Image Generation category.
No. Generation runs on OpenAI's servers and is reached through ChatGPT, the API or products built on top of it. There is no local installer, and every image depends on a network connection and available quota on your plan.
OpenAI's terms describe what is permitted, and they differ between the consumer products and the API. Read the current terms for your product before commercial use, and note that generated content may not be registrable as a trademark in some jurisdictions.
DALL-E trades control for convenience. It needs no hardware and handles prompt revision well, but you cannot fine-tune it on your own images, run it offline or swap in community checkpoints the way open-weight alternatives allow.
Industry-standard raster editor with generative AI features built in
Text-to-image generator known for its striking visual style
Open-weight image generation models that run on your own hardware
Community-built browser interface for running image models on your own machine