Google DeepMind has made its latest AI image generator, Imagen 3, free for use in the U.S. via the company’s ImageFX platform. Imagen 3 will be available in multiple versions, “each optimized for different types of tasks, from generating quick sketches to high-resolution images.” Google announced Imagen 3 at Google I/O in March, and in June made it available to enterprise users through Vertex. Using simplified natural language text input rather than “complex prompt engineering,” Google says Imagen 3 generates high-quality images in a range styles, from photorealistic, painterly and textured to whimsically cartoony.

Google says Imagen 3 is its highest quality text-to-image model, “capable of generating images with even better detail, richer lighting and fewer distracting artifacts than our previous models.”

PetaPixel tested it and says it “does appear to be a quality text-to-image model that rivals Midjourney or OpenAI’s DALL-E,” with the added benefit of being “currently free-to-use.”

“Google gives the option of editing the images using the now common inpainting technique,” notes PetaPixel, explaining “this method allows the user to select a part of the image and type in the change they would like to see.”

To help Imagen 3 capture nuances “like specific camera angles or compositions in long, complex prompts,” Google DeepMind “added richer detail to the caption of each image in its training data,” resulting in “better information to learn from” and more accurate generation across subject matter and styles, the company explains.

PetaPixel notes “there is little word on the data used to train Imagen 3,” citing a research paper assertion that the model “was trained on a large dataset comprising images, text and associated annotations.”

Meanwhile, VentureBeat writes “many users report that seemingly innocuous prompts are being blocked” and compare that to xAI’s newly released Grok-2, which “offers image generation capabilities with virtually no restrictions.”