Developers can now integrate image generator DALL-E 2 directly into their apps and products using the beta API released this week by OpenAI.
DALL-E 2 creates art by interpreting words you enter into a prompt (with filters for hate symbols and gore). More than 3 million people are already using the program, which currently produces 4 million-plus visuals per day. Now, "developers can start building with this same technology in a matter of minutes," OpenAI says.
Developers can access OpenAI's DALL-E 2 API online. Pricing varies by resolution: 1,024-by-1,024 images cost $0.02 each; 512-by-512 are $0.018 each; and 256-by-256 are priced at $0.016 each. Volume discounts are available to companies working with OpenAI's enterprise team.
"As our research evolves, we will continue to bring the state of the art into the API, including advances in image quality, latency, scalability, and usability," OpenAI says.
The research lab has already teamed up with Microsoft, which is building DALL-E into its new Designer app for creating professional-quality graphics, as well as the Bing search engine and Edge web browser—so users can draft their own images when the internet lets them down.
Other early adopters include fashion and lifestyle operating system CALA, whose DALL-E-powered AI tools let folks generate new design ideas from natural text descriptions or uploaded reference images. Photo startup Mixtiles also relies on the DALL-E API to create and frame "emotionally resonating artwork" by guiding customers through "a creative process that captures childhood memories, dream destinations, and more."