Tuesday, April 29, 2025

OpenAI Introduces Advanced Image Generation in ChatGPT, Faces GPU Overload

OpenAI has revealed a significant new feature for ChatGPT with the introduction of advanced image generation capabilities based on its GPT-4o model. The new feature allows users to create visually pleasing and practical images within the ChatGPT interface. Unlike the previous application of DALL-E, the GPT-4o model offers native multimodal functionality, enabling precise, photorealistic outputs that can represent text accurately and react to complex prompts.

The GPT-4o image generator is designed for everyday use, such as generating logos, diagrams, infographics, and other everyday images. It is capable of multi-turn generation, allowing users to refine and perfect images while maintaining consistency. It is also able to handle up to 20 different objects within a single prompt, an increase from earlier models. Users can also post images to edit or use as a reference for fresh work. The latest text rendering of the model provides smooth integration of text content into images, supporting visual communication.

This feature is now being rolled out to ChatGPT Plus, Pro, Team, and Free users. Enterprise and education customers will have access in the coming weeks. Developers will soon be able to use the GPT-4o API to create images. Users can customize their outputs by giving aspect ratios, colors through hex codes, or even transparent backgrounds.

OpenAI has incorporated C2PA metadata in images produced to enable accurate identification by AI image detectors. The platform also includes measures for not producing hazardous content, such as child sex abuse material or sexual deepfakes. OpenAI has expanded restrictions on imagery production involving real people and included robust safeguards against nudity and graphic violence.

Due to phenomenal demand, OpenAI briefly paused free-tier customers’ access but will resume rollout soon. The company acknowledges existing constraints in the model and is committed to ongoing enhancement. This upgrade is an improvement in AI-generated images with a balance of imagination and usefulness for widespread applications.

- Advertisment -
Google search engine

Most Popular