Release2026-08-08

xAI released Imagine Image 2.0 as the new Quality Mode in Grok's Imagine image service on the web and in its iOS and Android apps. The model targets work needing controlled layouts, readable text in images and repeated visual elements, and adds tools for editing a selected region, removing backgrounds and resizing to a chosen frame. There is no programming interface yet, so it can only be used inside the apps.

What changed

Grok's earlier image model scored 1228 for text-to-image and offered no region-targeted editing or resizing tools.

What it unlocks

Editing one selected area of an image while leaving the rest untouched, extending an image to a different frame shape, and combining up to five reference pictures so recurring characters and objects stay consistent.

  • up to 5 reference images per generation
  • 1320 text-to-image Arena score vs gpt-image-2 at 1380
  • 1439 editing score vs gpt-image-2 at 1463
  • previous Grok image model ranked #14, now #2

What you need to act on it

  • access to Grok Imagine on web, iOS or Android
  • no API access yet

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.