Everyone can now try Google's most advanced image generator

Google demoed Imagen 3, its most advanced text-to-image generator, at I/O this May. The tool recently made it to users as part of the Pixel Studio app on the Pixel 9 series, and you can even try it out with a Gemini Advanced subscription or Google’s AI Test Kitchen if you’re in the US. Over the coming days, Google will expand Imagen 3 availability to more users, and introduce Gemini Gems— a tool to customize Gemini and create AI experts on any topic.

Google claims that Imagen 3 sets a new standard for image quality and allows users to generate images in various styles with just a few words. The image generation model is capable of generating everything from photorealistic landscapes to textured oil paintings. The company previously asserted that it emphasizes delivering images with better detail, richer lighting, and fewer artifacts compared to other image generation models.

So far, Imagen 3 has been limited to Gemini Advanced subscribers in the US and only supports English. However, Google will expand its availability to all Gemini apps and add support for all languages over the next few days.

Along with wider Imagen 3 availability, Google has announced that it will start rolling out an early-access version of the tool that will be able to generate images of people. Google has implemented a few safeguards for this version, which will prevent users from generating images of “photorealistic, identifiable individuals, depictions of minors, or excessively gory, violent or sexual scenes.”

The early-access version will initially be available for Gemini Advanced, Business, and Enterprise users in English over the next few days.