OpenAI CEO Sam Altman has announced a major upgrade to its popular search engine, ChatGPT’s image-generation capabilities, which marks the first significant enhancement in over a year. The latest feature will be called ‘Images in ChatGPT’, which enables users to create high-quality images directly within the chatbot.
Available across ChatGPT Plus, Pro, Team, and Free plans, the new update has been introduced as OpenAI’s most advanced image-generation model, GPT-4o, which now replaces DALL-E 3 as the platform’s primary AI-powered image creator.
How GPT-4o improves image generation
Unlike DALL-E 3, which functioned as a separate diffusion transformer model, GPT-4o is an all-in-one AI system trained to handle text, images, and code within the same model. This unified approach enables the AI to produce images with:
- Greater accuracy and detail
- Better consistency across multiple iterations
- Enhanced contextual understanding of prompts
Altman explained that while GPT-4o takes slightly longer to generate images, the extra processing time results in more precise and realistic visuals.
New features and capabilities
According to OpenAI, GPT-4o’s image model was trained on a vast dataset of text and images, allowing it to understand the intricate relationships between language and visuals. Some key improvements include:
- Interactive Image Editing – Users can refine images dynamically through real-time chat, adjusting elements like backgrounds and objects with simple prompts.
- Context-Aware Modifications – GPT-4o can edit existing images, including those with people, seamlessly integrating adjustments to foreground and background elements.
- Improved Object Handling – While other AI systems struggle with rendering 5-8 objects accurately in a scene, GPT-4o can handle 10-20 with improved spatial awareness and composition.
For instance, if you're designing a video game character, GPT-4o ensures that the character’s design remains consistent across multiple refinements, offering a more controlled creative process.
A smarter AI for image and text creation
By deeply integrating text and visual understanding, GPT-4o enhances ChatGPT’s overall capabilities, making it not just a better image creator but also a more intelligent assistant for designers, content creators, and developers.
ALSO READ: Microsoft Windows 11 25H2 testing begins: What to expect?
ALSO READ: OnePlus 13T set to launch with 6,200mAh battery, resembles iPhone 16: What to expect?