OpenAI released ChatGPT Images 2.0, a new image generator. The company claimed that this new model “can take on complex visual tasks and produce precise, immediately usable visuals, with sharper editing, richer layouts, and thinking-level intelligence.”
In a tweet, the company shared a “video made with ChatGPT Images”, and the pictures look incredibly realistic – to the point that some may find it hard to distinguish the AI-generated images from the real ones.
How to use it?
Here is a quick guide to accessing the new AI model, which OpenAI says is available to all ChatGPT and Codex users.
1. Visit ChatGPT Images 2.0
2. Click on the “Try in ChatGPT” option
Also Read: 6 straightforward prompts to try on Google Ask Maps to improve your search results
3. A new page will open with several options that you can take inspiration from. They include several styles from headshots to comics to 3D avatars to anime and more.
4. You can click on a style, upload any image you want to refine and let ChatGPT do its work
5. Alternatively, you can also “Describe a New Image” and create it using the AI tool
ChatGPT Images 2.0 prompts:
HT.com asked ChatGPT to share a list of the most effective prompts for its latest model. In response, the AI chatbot provided a diverse range of suggestions tailored to various categories, such as hyper-realistic cinematography, conceptual art, environmental design, and professional photography.“
Here are some of the prompts you can use.
#Prompt 1
“Overhead food photography of a rich Indian thali, vibrant colours, natural lighting, high texture detail”
#Prompt 2
“Old monk meditating in a mountain temple, soft sunlight, peaceful aura, cinematic composition”
#Prompt 3
“Post-apocalyptic Indian metro city, cracked roads, nature reclaiming buildings, warm dusty sunlight, ultra-realistic”
#Prompt 4
“A person drowning in social media icons, dark tone, commentary on digital overload, surreal realism”
#Prompt 5
“Rain-soaked city street at night, neon reflections, slow shutter motion blur, pedestrians with umbrellas, cyberpunk tone, cinematic lighting”
How did social media react?
An individual wrote, “I think the real test is whether small business owners can use this without a designer. In my experience building products, the bottleneck is rarely the idea; it’s the asset production. If Images 2.0 removes that dependency, it’s genuinely useful for founders.” Another added, “Loving how it kept the lighting consistent across every object change. Small detail, big difference.”
A third commented, “I have just tried it, and it is amazing.” A fourth wrote, “Bro, the precision on this new image model is insane. Dense layouts, accurate text, flexible ratios… and it actually understands what you want now. That promo video they made with it is a chef’s kiss.
About ChatGPT Images 2.0:
OpenAI described the new AI model as a “new era of image generation.”
Also Read: Gemini AI saree photos: Red saree prompts to turn your selfie into a vintage poster
“To extend the model’s capabilities for the most complex tasks, Images 2.0 is our first image model with thinking capabilities. When a thinking model is selected in ChatGPT, Images 2.0 can search the web for real-time information, create multiple distinct images from one prompt, and double-check its own outputs. With thinking, the model can take on even more of the heavy lifting between idea and image, especially when accuracy, up-to-date information, consistency, and visual cohesion matter most,” it further stated.
