Introducing MOSS-Image technology, which allows you to arrange the positions of objects and text in AI-generated images simply by using text and layout inputs.
Explore a new technology that bypasses the traditional VAE, the standard for AI image generation, by creating images directly in the Visual Foundation Model (VFM) space.
We explore the speed improvements, new Sketch feature, and precision editing capabilities of ChatGPT Images 2.5, the latest image generation model from OpenAI.
An easy-to-understand introduction to DiffusionBench, a unified platform for systematically measuring the performance of AI image generation models.
Google's latest AI, Gemini 2.0 Flash, can now generate and edit images through simple conversation. We explain what 'native' image generation is and why it matters in easy-to-understand terms.