DALL·E 2

Free

Announcement of the release of DALL·E 2, an advanced image generation system with improved resolution, expanded image creation capabilities, and various safety mitigations. OpenAI blog, April 6, 2022.

FreeFree tier

Inputs: text, imageOutputs: image

Type

Open Source

Company

OpenAI

Visit Website GitHub (40 stars)

About DALL·E 2

DALL·E 2 is an AI system by OpenAI that generates realistic images and art from natural language descriptions. It offers 4x higher resolution than its predecessor DALL·E, combines concepts, attributes, and styles, and includes safety mitigations such as limiting violent or adult content and preventing photorealistic faces of real individuals. Key capabilities include creating original images from text, editing existing images with inpainting, and integration with ChatGPT. The system was released in beta in July 2022 and is available through OpenAI's platform and the ChatGPT interface.

Key Features

Generates original, realistic images and art from natural language descriptions

Combines concepts, attributes, and styles in generated images

4x greater resolution than DALL·E 1

Preferred over DALL·E 1 for caption matching (71.7%) and photorealism (88.8%)

Safety mitigations: prevents violent, hate, or adult images; limits photorealistic faces of real individuals

Phased deployment with automated and human monitoring to curb misuse

Available in beta, integrated with ChatGPT for chat-based image generation

Supports inpainting (editing images via text prompts) and outpainting (extending images beyond original boundaries)

Pros & Cons

Pros

Produces high-resolution, realistic images from text
Ability to combine diverse concepts, attributes, and artistic styles
Strong safety measures to reduce harmful content generation
Integration with ChatGPT for accessible conversational image generation
Continuous improvements through phased deployment and user feedback

Cons

Limited availability in beta with potential access restrictions
Content filters may limit creative expression for certain topics
Cannot generate photorealistic faces of real individuals (including public figures)
Requires internet connection and access to OpenAI's platform
Usage costs may apply via ChatGPT subscription or API credits (not free after initial credits)

Best For

Creating art and illustrations from textual descriptionsGenerating photorealistic images for design and marketingExploring creative concepts and visual brainstormingRapid prototyping of visual ideas for products or presentationsEditing existing images with text-based modifications (inpainting)

FAQ

What is DALL·E 2?

DALL·E 2 is an AI system by OpenAI that creates realistic images and art from natural language descriptions. It can combine concepts, attributes, and styles, and offers 4x higher resolution than its predecessor.

How does DALL·E 2 differ from DALL·E 1?

DALL·E 2 generates more realistic and accurate images with 4x greater resolution. In evaluations, it was preferred over DALL·E 1 for caption matching (71.7%) and photorealism (88.8%).

What safety measures are in place?

OpenAI has implemented filters to prevent violent, hate, or adult images, limited generation of real faces, and uses automated and human monitoring systems to guard against misuse.

Where can I use DALL·E 2?

DALL·E 2 is available in beta through OpenAI's platform and is integrated with ChatGPT for chat-based image generation. It can also be accessed via the DALL·E API.