xAI releases Grok Imagine Image 2.0, ranks second on Arena
xAI released Grok Imagine Image 2.0 on August 8 with stronger image editing capabilities, ranking second on Arena for text-to-image and image editing.
On August 8, xAI released Grok Imagine Image 2.0, the company's new flagship image generation model. The model is rolling out as a new quality mode on the Grok web app, iOS and Android apps, with a developer API available in parallel.
Image 2.0 focuses on professional image editing with localized modification and asset reuse as the core optimizations. It targets commercial design and product photography workflows. Instruction following has been substantially improved, allowing the model to autonomously plan layout and text composition for complex multi-element images while keeping small-size text readable and preserving subject identity across multi-turn iterations.
The model includes a magic-wand local editing tool that lets users modify a selected region without affecting other elements, a precise selection tool for local recoloring and asset replacement, and a background remover that can export transparent assets. It accepts up to five reference images per request and offers intelligent aspect-ratio adaptation across nine common ratios from 1:2 to 2:1.
As of the August 7 Arena snapshot, Grok Imagine Image 2.0 ranks second globally in both text-to-image and image editing overall scores, behind only OpenAI's GPT Image 2. The model was fine-tuned for photography, graphic design, and illustration, with improvements to lighting fidelity, material detail, and color consistency.