Imagen 4

June 26, 2011 - Google has launched the latest version of its text-generating image model, Imagen 4, with a high-end version, Imagen 4 Ultra, an update designed to significantly improve text rendering compared to its predecessor, Imagen 3. Both versions are currently available in a paid preview of the Gemini API and in a limited free beta at Google AI Studio. Google is positioning Imagen 4 as a general-purpose model "for most tasks," with pricing per image ...

Google DeepMind's Imagen 4 AI model generates richer, more detailed images faster than ever before, with superior typography and color, up to 2k resolution, and improved adherence to cues. Available now on Gemini, Whisk, and Vertex AI, Imagen 4 is an advanced AI image generation model that generates high-quality images based on textual cues. With photo-realism, a wide range of artistic styles, and ultra-fast generation speed, it is suitable for a variety of creative scenarios, such as advertising design, game development...