Google is rolling out a brand new image-generating AI mannequin, Imagen 4, that the corporate claims delivers higher-quality outcomes than its earlier picture generator, Imagen 3.
Unveiled at Google I/O 2025 on Tuesday, Imagen 4 is able to rendering “fantastic particulars” like materials, water droplets, and animal fur, Google says. The mannequin can deal with each photorealistic and summary kinds, creating photos in a variety of facet ratios and as much as 2K decision.
“Imagen 4 is [a] big step ahead in high quality,” Josh Woodward, who leads Google’s Labs group, mentioned throughout a press briefing. “We’ve additionally [paid] a whole lot of consideration and fixes round the way it generates textual content and topography, so it’s great for creating slides or invites, or some other factor the place you may have to mix imagery and textual content.”
There’s no scarcity of AI picture mills on the market, from ChatGPT’s viral software to Midjourney’s V7. They’re all comparatively subtle, customizable, and able to creating high-quality AI art work. So what makes Imagen 4 stand out from the group?
Based on Google, Imagen 4 is quick — sooner than Imagen 3. And it’ll quickly get sooner. Within the close to future, Google plans to launch a variant of Imagen 4 that’s as much as 10x faster than Imagen 3.
Imagen 4 is accessible as of this morning within the Gemini app, Google’s Whisk and Vertex AI platforms, and throughout Google Slides, Vids, Docs, and extra in Google Workspace.
