Open AI’s DALL-E 2 AI image generation model is no longer cutting-edge.
Today, the company announced DALL-E 3, its latest text-to-image generator and showed off some of its new impressive features, including the ability to generate readable text baked directly into images themselves — something that was not easy with DALL-E 2, and which other competing image generator AI models such as Midjourney still struggle to achieve.
“DALL·E 3 delivers significant improvements over DALL·E 2 when generating text within an image and in human details like hands,” OpenAI wrote on its web page explaining the new model.
This feature puts OpenAI in direct competition with Ideogram, a startup from former Googlers launched last month, which also offers image generation with text/typography baked in using its own proprietary AI model.
Furthermore, OpenAI wrote that DALL-E 3 does a much better job of understanding the spatial relationships that users include in their prompt text, generating imagery that places figures and objects where the user has described in relation to one another. This means that descriptive prompts can now be rendered far more accurately, as seen in an example screenshot below.
OpenAI also said that DALL-E 3 would be coming to ChatGPT Plus, the paid $20-per-month subscription tier of its hit large language model (LLM), and its new ChatGPT for Enterprise plans announced last month, meaning that corporate clients will now have the ability to generate imagery with text for their marketing or internal collateral.
In addition, OpenAI says that ChatGPT can help users refine their prompts automatically to generate the imagery that better matches their intent.
A video posted by OpenAI co-founder and CEO Sam Altman on X, the social network formerly known as Twitter, demonstrates the impressive back-and-forth conversational prompting style that is now possible in DALL-E 3 thanks to the ChatGPT integration.
also, the video we made for dalle 3 is SO CUTE: pic.twitter.com/k1FOFTOsU5
At the same time, OpenAI wrote that “like previous versions, we’ve taken steps to limit DALL-E 3’s ability to generate violent, adult, or hateful content.”
The announcement was cheered on by OpenAI developer relations advocate Logan Kilpatrick on X (formerly Twitter), who said it was “absolutely incredible.”
Huge news: @OpenAI DALL-E 3 will soon available in ChatGPT Plus and ChatGPT Enterprise ?
This latest DALL-E model is absolutely incredible, I have been blown away by what it is able to generate. pic.twitter.com/eTWzxiOHgB
Head over to our on-demand library to view sessions from VB Transform 2023. Register Here
Open AI’s DALL-E 2 AI image generation model is no longer cutting-edge.
Today, the company announced DALL-E 3, its latest text-to-image generator and showed off some of its new impressive features, including the ability to generate readable text baked directly into images themselves — something that was not easy with DALL-E 2, and which other competing image generator AI models such as Midjourney still struggle to achieve.
“DALL·E 3 delivers significant improvements over DALL·E 2 when generating text within an image and in human details like hands,” OpenAI wrote on its web page explaining the new model.
This feature puts OpenAI in direct competition with Ideogram, a startup from former Googlers launched last month, which also offers image generation with text/typography baked in using its own proprietary AI model.
Event
VB Transform 2023 On-Demand
Did you miss a session from VB Transform 2023? Register to access the on-demand library for all of our featured sessions.
Understands spatial relationships
Furthermore, OpenAI wrote that DALL-E 3 does a much better job of understanding the spatial relationships that users include in their prompt text, generating imagery that places figures and objects where the user has described in relation to one another. This means that descriptive prompts can now be rendered far more accurately, as seen in an example screenshot below.
Integrated with ChatGPT
OpenAI also said that DALL-E 3 would be coming to ChatGPT Plus, the paid $20-per-month subscription tier of its hit large language model (LLM), and its new ChatGPT for Enterprise plans announced last month, meaning that corporate clients will now have the ability to generate imagery with text for their marketing or internal collateral.
In addition, OpenAI says that ChatGPT can help users refine their prompts automatically to generate the imagery that better matches their intent.
A video posted by OpenAI co-founder and CEO Sam Altman on X, the social network formerly known as Twitter, demonstrates the impressive back-and-forth conversational prompting style that is now possible in DALL-E 3 thanks to the ChatGPT integration.
At the same time, OpenAI wrote that “like previous versions, we’ve taken steps to limit DALL-E 3’s ability to generate violent, adult, or hateful content.”
The announcement was cheered on by OpenAI developer relations advocate Logan Kilpatrick on X (formerly Twitter), who said it was “absolutely incredible.”
VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.
Author: Carl Franzen
Source: Venturebeat
Reviewed By: Editorial Team
