Google has released the latest version of Imagen 3, its text-based AI image generator, to users in the United States. Accessible through Google’s AI Test Kitchen platform, it promises to generate images with “better detail, richer lighting, and fewer artifacts” than previous models Google developed. The company first unveiled the updated Imagen 3 at Google I/O in May, but the tool has only recently been made widely available through its Vertex AI platform. In recent days, some Reddit users have begun experimenting with Imagen 3, and on Tuesday, Google published a detailed research paper on the tool.
Like other generative AI image generators, Imagen 3 can create detailed images from simple text. Additionally, you can modify the generated image by highlighting a specific area and describing the desired change. It appears to have some security measures in place: the tool refuses to generate images of public figures and does not produce images of weapons. While it avoids creating copyrighted characters, you can still work around this limitation by describing the desired character specifically.
#Google #updates #AIpowered #image #generator
