OpenAI releases new image generation model with complex capabilities
ChatGPT Images 2.0 marks a big leap forward in image generation as OpenAI seeks to distinguish its features from Anthropic’s Claude.
OpenAI just released a major upgrade to ChatGPT’s image generation capabilities with ChatGPT Images 2.0. Powered by its new GPT Image 2 model, the image generation tool operates similar to earlier image generation features, turning a text prompt into an image. But the new tool goes a step further, allowing more varied styles and more complex instructions by performing research to synthesize into the final output, much like ChatGPT’s reasoning and “thinking” features.
Users can generate multiple images from one prompt, and can produce very realistic images that don’t look as AI-generated as previous versions.
The tool can generate images at different aspect ratios and can assemble complex design layouts and infographics. Text generation has been also improved, which has traditionally been a weak spot for AI image generators.
Last week, Anthropic released Claude Design, a tool that can also generate complex layouts, designs, and interfaces, which tanked shares of Figma. OpenAI’s announcement did not move shares of Figma, or Adobe.
