AI company OpenAI has introduced new image generation capabilities within its GPT-4o model, enabling the production of photorealistic visuals with greater precision.
The company announced that the upgraded model can now render text more accurately, follow complex instructions, and handle up to 20 distinct objects within a single image.
Go deeper with GlobalData
In a blog post, OpenAI said the model’s ability to merge symbols seamlessly with imagery and engage in multiturn image generation, allows users to refine and iterate on creations through natural conversation.
The new features also allow GPT-4o to learn from user-uploaded images, enhancing the contextual accuracy of its outputs.
“GPT-4o image generation excels at accurately rendering text, precisely following prompts, and leveraging 4o’s inherent knowledge base and chat context—including transforming uploaded images or using them as visual inspiration,” the company said.
“These capabilities make it easier to create exactly the image you envision, helping you communicate more effectively through visuals and advancing image generation into a practical tool with precision and power.”
OpenAI confirmed that all images generated by GPT-4o will carry C2PA metadata to identify them as AI-generated.
The company also reiterated its commitment to content safety by continuing to block image-generation requests that violate its policies.
The rollout of GPT-4o’s image generation feature has begun for Plus, Pro, Team, and Free users, with availability for Enterprise and Edu customers set to follow.
According to OpenAI, “Creating and customising images is as simple as chatting using GPT-4o — just describe what you need, including any specifics like aspect ratio, exact colours using hex codes, or a transparent background.”
GPT-4o was initially launched in May 2024, offering text, vision, and audio processing improvements.
The model is available for free, with paid subscribers receiving increased usage limits. In August, OpenAI introduced fine-tuning capabilities for GPT-4o, allowing developers to adapt the model for specific business needs.
In February 2025, OpenAI announced the expansion of its AI agent, Operator, to more regions across the globe.
In a post on X, the company said: “Operator is now rolling out to ChatGPT Pro subscribers in Australia, Brazil, Canada, India, Japan, Singapore, South Korea and the UK.”
