OpenAI introduced the DALL-E 3 image generator. It will arrive to ChatGPT Plus in October

This generation promises more natural and intuitive prompting, eliminating the need for intricate prompt engineering and allowing for smoother user interaction through ChatGPT.

By Veronica Marshall

Sep 22, 2023, 09:45 PM GMT+0

michael90212 image gallery in the museum with pop art pictures 72d56dfe-df1c-48f3-9e3a-8513f959572b

OpenAI has introduced the third generation of the DALL-E image AI generator, DALL-E 3. Previously, although the technology was advanced, it was not converted into an equally accessible and high-quality service like its competitor, Midjourney.

The upcoming release will be available as an exclusive feature for ChatGPT Plus subscribers, enterprise clients, and developers starting October.

The new generation is expected to be more creative and intuitive in interpreting full text commands, allowing for more user-friendly interactions and more effective output. This is aimed at removing the necessity for the often complex and specifically structured text commands, known as prompt engineering, which are currently needed to make the generator produce the desired result.

DALL-E 3 is designed to consider all the information provided in the text command during the generation process. Additionally, the prompting is expected to be much more user-friendly and intuitive.

The OpenAI team has provided initial examples of what DALL-E 3 is capable of on their website, and more can be found on Instagram or on Sam Altman's Twitter account, CEO of OpenAI.

Some of these examples include prompts added to the images. Comparisons have been made with Midjourney by inputting the same prompts to observe differences in outputs.

DALL-E 3 Prompt: An ink sketch style illustration of a small hedgehog holding a piece of watermelon with its tiny paws, taking little bites with its eyes closed in delight.

Midjourney 5.2 prompt: An ink sketch style illustration of a small hedgehog holding a piece of watermelon with its tiny paws, taking little bites with its eyes closed in delight --s 250

Midjourney 4 prompt: An ink sketch style illustration of a small hedgehog holding a piece of watermelon with its tiny paws, taking little bites with its eyes closed in delight --s 250 --v 4

It is apparent from the presented images that DALL-E 3 interacts effectively with text commands. In comparison, making Midjourney add captions is often impossible. OpenAI seems to focus on delivering precise real images in response to the given prompts. The specific details of DALL-E 3's capabilities are yet to be fully disclosed, but OpenAI assures enhanced creativity and improved attention to the text commands provided.