Grok Imagine 2.0 Is xAI’s New Shot at Dominating AI Images

Image 2.0 ranks second globally in both text-to-image generation and image editing, based on Arena's leaderboards as of August 7.

Grok Imagine 2.0 Is xAI’s New Shot at Dominating AI Images

xAI has released Imagine Image 2.0, a new image generation and editing model designed to produce visuals for professional creative work rather than just one-off AI experiments.

The model is now generally available as Quality Mode on Grok.com and through Grok's iOS and Android apps.

xAI says Image 2.0 is built to follow detailed instructions more accurately, including typography, layout and complex compositions, while maintaining consistency across multiple generations and edits.

"We built Image 2.0 around a simple goal: make images you can use in real work. 2.0 follows instructions closely, down to the details. It plans typography and layout the way a designer would, so dense, multi-part visuals hold together and small text comes out sharp. And it preserves what you put in across generations and edits," the company said.

Editing is a major focus of the release. Users can selectively modify specific regions using a magic wand, isolate portions of an image through segmentation and remove backgrounds to create transparent assets.

The model also supports multi-reference editing, allowing users to provide up to five images in a single generation.

A new Smart Resize feature allows users to change an image's aspect ratio while having the model intelligently extend the scene to fill the new frame.

xAI is also introducing templates for common creative workflows, including product photography, editorial posters, professional headshots, icons, character sprites, marketing assets and e-commerce images.

According to xAI, Image 2.0 ranks second globally in both text-to-image generation and image editing, based on Arena's leaderboards as of August 7. OpenAI's GPT Image 2 currently ranks first in both categories.

The company is also positioning Image 2.0 for video workflows, allowing users to create consistent characters, locations and props across separate generations. API access is planned for a later date.

Last month, xAI open-sourced Grok Build, its AI-powered coding agent and terminal user interface (TUI), making the source code publicly available on GitHub to encourage transparency, community contributions, and broader adoption.