DALL·E 3

DALL·E 3 is OpenAI's third-generation text-to-image generation model that converts natural language descriptions into high-quality images. It is available to all ChatGPT users through a conversational interface and to developers via the OpenAI API. DALL·E 3 is designed for content creators, designers, marketers, developers, and anyone who needs to generate custom visuals from text prompts.

How It Works

DALL·E 3 is built natively on ChatGPT, which allows users to use ChatGPT as a brainstorming partner and prompt refiner. Instead of writing complex prompt engineering formulas, users can describe what they want to see in plain language, from a simple sentence to a detailed paragraph. ChatGPT automatically generates tailored, detailed prompts for DALL·E 3 to bring the idea to life. If an image is not quite right, users can ask ChatGPT to make tweaks with just a few words. For developers, DALL·E 3 is accessible through the OpenAI API for integration into applications and workflows.

Key Capabilities

DALL·E 3 represents a significant leap forward in prompt adherence compared to DALL·E 2. Modern text-to-image systems often ignore words or descriptions, forcing users to learn prompt engineering. DALL·E 3 generates images that more accurately follow the text provided, even with the same prompts that DALL·E 2 struggled with. The model can produce images across a wide range of styles, including photorealistic renders, oil paintings, 3D renders, pixel art, flat design illustrations, ink sketches, and art deco posters. It handles complex scenes with multiple subjects, specific lighting conditions, artistic mediums, and compositional details.

Use Cases

DALL·E 3 serves a broad range of creative and professional applications. Content creators and marketers use it to generate custom visuals for blog posts, social media, and advertising campaigns without hiring illustrators or photographers. Designers use it for concept exploration, mood boarding, and rapid visual prototyping. Developers integrate it into applications via the API for features such as on-demand image generation, personalized visual content, and automated creative workflows. Educators and students use it for visual learning materials and creative projects. The model's ability to follow detailed prompts also makes it useful for generating specific product visualizations, storyboards, and artistic references.

Safety and Creative Control

DALL·E 3 includes several safety mitigations. The model declines requests that ask for an image in the style of a living artist and declines requests that ask for a public figure by name. OpenAI partnered with red teamers to improve safety performance in areas such as harmful biases related to visual over-representation and under-representation, propaganda, and misinformation. Creators can opt their images out from training of future image generation models. OpenAI is also researching provenance classifiers to help identify whether an image was generated by DALL·E 3, aiming to help people identify AI-generated content.

Availability and Pricing

DALL·E 3 is available to all ChatGPT users through the ChatGPT interface. ChatGPT Plus, Team, Enterprise, and Edu subscribers can generate images directly in chat conversations. For developers, DALL·E 3 is accessible via the OpenAI API on a pay-per-use basis, with pricing based on image resolution. As with DALL·E 2, the images created with DALL·E 3 are owned by the user, and no permission from OpenAI is needed to reprint, sell, or merchandise them.