
OpenAI has recently integrated its default image generation tool directly into the GPT-4o model, enabling users to create images within ChatGPT through simple prompts—eliminating the need to rely on DALL·E as a separate interface.
Beyond generating visuals through text prompts, conversational cues, or uploaded files, users can now refine and modify previously created images using additional instructions. The system also boasts significantly improved comprehension of textual descriptions, allowing for greater visual consistency between successive generations, thereby avoiding stark discrepancies.
This newly embedded feature is designed to cater to both personal and professional use cases, including the creation of infographics, social media content, and materials rich in visual elements and explanatory graphics.
Initially, access to this tool will be granted to paid subscribers of ChatGPT Plus, Pro, and Team plans. However, OpenAI has announced plans to extend availability to free-tier users in the future, with further integration into the creative platform Sora also on the horizon.