xAI's Imagine Image 2.0 ranks #2 in Arena benchmarks, behind OpenAI

xAI has released Imagine Image 2.0, a new image generation and editing model available immediately as a Quality Mode on grok.com/imagine and within Grok's iOS and Android applications. API access for the model will arrive at a later date. The company describes the model as capable of following instructions with fine-grained accuracy, maintaining clean typography and layout in complex visuals, and sustaining visual consistency across multiple generations.

According to Arena leaderboards dated August 7, 2026, Imagine Image 2.0's faster "low" variant ranks in second place globally across both major evaluation categories. In the Image Edit Arena, it achieves an Elo rating of 1,439, trailing OpenAI's GPT-Image-2 at 1,463. In the Text-to-Image Arena, it scores 1,320, again behind GPT-Image-2's 1,380. Competitors including Reve 2.1, Meta's Muse-Image, Alibaba's Qwen-Image-3.0-Pro, Google's Gemini, and ByteDance's SeedDream rank further down the leaderboards. The new model significantly outperforms the "Quality" variant of its predecessor in these benchmarks.

Imagine Image 2.0 ships with several editing tools designed for iterative creative work. A feature called Magic Wand modifies only selected areas of an image, while a segmentation tool lets users pick precise regions. A background removal tool exports subjects with transparent backgrounds. Multi-Ref Editing allows combining up to five input images into a single generation. Smart Resize converts existing images to any aspect ratio, with the model automatically filling in additional space.

xAI is also introducing templates that bundle common workflows into preconfigured starting points across categories including photo editing, product photography, marketing materials, design tools, game assets, and streaming emojis. The company is showcasing a feature that generates characters, locations, and props separately while maintaining consistent visual style across all images—a capability xAI describes as a foundation for future full video production workflows. According to xAI, Imagine 2.0 can generate consistent visual worlds with characters, props, and locations that serve as a starting point for video production.