In a surprising announcement, OpenAI has revealed the upcoming release of Dall-E 3, the latest iteration of their text-to-image generation model. Set to be in research preview, Dall-E 3 promises to redefine the way we create images from text descriptions.
This new version of Dall-E is set to become available to ChatGPT Plus and Enterprise customers in October through the API, with a later release in Labs during the fall.
Perhaps the biggest implementation is that this version of Dall-E will be seamlessly integrated into ChatGPT, meaning you can use this tool to generate images with quick, concise prompts, giving Dall-E a big leg up over other competitors in the AI image-generating field.
The key challenge OpenAI has tackled is that modern text-to-image systems have the tendency to misinterpret or disregard specific words or descriptions, necessitating intricate prompt engineering by users. However, Dall-E 3 heralds a significant advancement in addressing this issue by ensuring that the generated images closely align with the provided text input.
That feature is what sets Dall-E 3 apart from previous versions of this tool, with is its adaptability and user-friendly nature also improved. If a generated image isn't quite perfect, users can prompt ChatGPT to make swift adjustments with just a few words, offering an unprecedented level of control and fine-tuning.
As mentioned, Dall-E 3 will be available to ChatGPT Plus and Enterprise customers starting in early October, continuing the model's tradition of granting users full rights to the images they create. Users are not required to seek OpenAI's permission for reprints, sales, or merchandise featuring these images.
A social media tool emphasises how Dall-E 3 will work in practice, take a look at the video below.
Will you be using this feature for ChatGPT? If you want to see if this AI tool is for you, we made a guide explaining if ChatGPT Plus is worth the subscription or if you're better off simply using the free tools available.