During I/O, Google introduces a new version of its image generator Imagen 4, which should offer “more details” and “better typography”.
Imagen 4 “combines speed with precision to create stunning images”, according to Eli Collins, VP of product at Google Deepmind, in a blog post. This fourth generation of image generator, the successor to Imagen 3, is also said to be better at spelling and typography. Imagen 4 is now available in the Gemini app, Whisk, Vertex AI, and in Slides, Vids, Docs “and more in Workspace”.
Details and Typography
According to the company, the latest model creates “stunning images” with “remarkable clarity in fine details”. Google demonstrates this with several example images, such as the whale below.
Furthermore, Imagen 4 can create images in various aspect ratios and a resolution up to 2K, which is even better for prints or presentations. Additionally, Google states that the model “is significantly better at spelling and typography”. This means that Imagen 4 can much more easily create greeting cards, posters, or even comics, without strange spelling or typography.
Imagen 4 is available from today in the Gemini app, Whisk, Vertex AI, and in Slides, Vids, Docs, and more in Workspace. Furthermore, Google aims to launch a fast variant of Imagen 4 “soon” that is up to 10x faster than Imagen 3.