Imagen 4 is Google’s newest and strongest text-to-image model. They showed it off at Google I/O on May 20 2025. It makes super clear photoreal images that get the tiny stuff right like fine fabric patterns water drops and animal fur.
It’s also good with text inside the images. So it actually writes words you can read which is perfect for things like posters or UI mockups.
Sharp image control. It works with five image shapes. You got square (1:1) portrait (9:16) landscape (16:9) mobile portrait (3:4) and mobile landscape (4:3).
Native image sizes. Each shape has its own size baked in. 1024x1024 for square 1408x768 for 16:9 1280x896 for 4:3 896x1280 for 3:4 and 768x1408 for 9:16.
Talk to it in your language. You can write prompts in English Chinese Hindi Japanese Korean Portuguese or Spanish.
Built-in watermark. It tags its own images using SynthID so people can tell it came from AI.
Fast results. It’s already quick but there's a faster Imagen 4 version coming that’s up to 10x speedier than Imagen 3.
If you'd like to access this model, you can explore the following possibilities:
Use our video cost calculator to compare prices between platforms offering Imagen 4 model.
For locally hosted models, see description and additional links at the bottom for versions, repos and tutorials.
