Grok Imagine Image 2.0 is the image model behind the Quality Mode in Grok Imagine on web, iOS, and Android, and is also exposed to developers through the Imagine API under the model identifier grok-imagine-image-2.0. The release is positioned around usable creative assets rather than one-shot novelty, with the model designed to plan typography and layout, follow detailed instructions, and preserve supplied elements across edits. It targets photography, graphic design, and illustration work, and the consumer surface adds practical tools such as a magic wand for localized edits, segmentation for exact region selection, transparent-background removal, smart resize, and templates for common jobs like product shots, headshots, collages, icons, game assets, user-generated imagery, and merchandise. Multi-reference composition is supported, with reference limits varying by surface.

The model accepts text and image inputs and produces image outputs, and developers can choose between 1K and 2K resolution at Low or Medium quality depending on their needs. API traffic is served from US East and US West, with rate limits starting at 6 requests per second and scaling upward for higher-tier accounts. Pricing is structured per image by resolution and quality tier, making the API a fit for production pipelines that need predictable cost per asset rather than experimentation. The combination of typography awareness, layout planning, localized editing, and reference-based composition makes it well suited for teams producing branded visuals, marketing content, and design assets where consistency, control, and resolution options matter more than raw generative novelty.