Hands-On with ChatGPT Images 2.0: A Significant Leap in AI-Generated Visuals
The latest iteration of ChatGPT's image generation engine, Images 2.0, showcases impressive capabilities in creating detailed text and graphics, making it a valuable tool for real-world applications.

Earlier this week, OpenAI unveiled ChatGPT Images 2.0, its latest image generation engine. A key highlight of this release is its ability to create full-page graphics, including detailed text, marking a significant improvement from its predecessor. I had early access to a pre-release version, which worked well but had issues with the ZDNET logo.
Now that the product has been officially released, I'm conducting an in-depth test across a wide range of challenges. Images 2.0 is available to all ChatGPT tiers, but the more advanced language features are exclusive to paying tiers that can use the Thinking model. I'm running these tests using a ChatGPT Plus account with Thinking turned on.
To begin, I focused on ZDNET branding exercises. Rather than uploading ZDNET pages and having it find the logo, I created a standalone image of the ZDNET logo and uploaded it with each prompt, which seemed to help tremendously. Note that ZDNET doesn't permit OpenAI to scrape its pages, as Ziff Davis, ZDNET's parent company, filed a lawsuit against OpenAI in April 2025 alleging copyright infringement.
I used a Chrome extension to capture full-screen screenshots of the articles I wanted to test with Images 2.0. My starting point was an article I previously wrote about Images 2.0. I fed ChatGPT the prompt: 'Create a detailed and vivid infographic of this article using the ZDNET brand style and the attached ZDNET logo.' The resulting image not only featured the correct logo but also perfect coloring for ZDNET.
Where the image truly excelled was in its use of text – all the text was correct, even the tiny text on an angle. Next, I revisited the sketchnotes challenge I gave to Google's Nano Banana a few months ago. The assignment was to create a sketchnote version of the US Bill of Rights.
While Nano Banana did a great job with the images, I had to try multiple times to get the wording right. For ChatGPT Images 2.0, I upped the stakes by requesting sketchnotes in ZDNET's branding style. The first prompt was: 'Make me a sketchnote of the US Bill of Rights.
Use the ZDNET logo style and make the sketchnotes in the ZDNET style.' The second prompt was: 'Include the ZDNET logo and add more neon-style colors, perhaps on a black background.' Both versions fit with ZDNET's style, and the text was correct with no duplicates or missing information. However, I noticed that the ZDNET logo looked jammed in on the second image. Despite this, the logo was correct, and a few more prompt passes could have improved its placement.
When I asked Images 2.0 to convert my AI website builder shootout article to an infographic, it produced a fairly usable, if somewhat busy, infographic. It even went online and added information I didn't have in the article, like base pricing. While these are errors an in-house human graphic designer might produce in a first draft, re-prompting Images 2.0 with corrections resulted in the correct modification of the infographic.
Source: ZDNet