Back in March, OpenAI announced a major upgrade to image generation in ChatGPT. Instead of relying on a separate model like DALL·E, ChatGPT began using GPT 4o native image generation capabilities to deliver more accurate and visually appealing images based on text prompts. The GPT 4o image generation model accurately renders text and follows prompts precisely by leveraging both its knowledge base and the ongoing chat context. It also enables users to edit uploaded images or generate entirely
