OpenAI launched DALL-E API

OpenAI has released the public beta version of the DALL-E API, which means developers can now integrate DALL-E directly into their applications and products.

With this announcement, DALL-E, a transformer language model that allows users to use natural language cues to create and edit source images, joins GPT-3, Embeddings and Codex on the OpenAI API platform.

The new API has already been implemented and tested in some products. For example, Cala, a fashion design platform, and Mixtiles, which prints online photos on lightweight decorative tiles, have already used DALL-E to create prints. And Microsoft is bringing DALL-E to its new Designer graphic design app, as well as integrating DALL-E in Bing and Microsoft Edge with Image Creator, allowing users to create images when web search results don’t turn up what they’re looking for. Image provider Shutterstock also announced last week that it will be using an API to serve DALL-E-generated images to customers.

Since the API is in beta testing, “we will continue to iterate and improve until the end of the year,” says OpenAI. “We’re very excited about all the ways that developers can use this technology and customize it for specific needs, specific applications, and specific communities to take the technology further than we’ve ever been able to.”

Open AI claims that 3 million people are already using DALL-E to boost creativity and speed up workflows, generating over 4 million images per day. Developers can now start building with DALL-E in minutes, they say.

The DALL-E API evaluates the number of output images depending on their size. 1024×1024 costs $0.02 per image, while there are very small discounts for 512×512 ($0.018) and 256×256 ($0.016) images.