Imagine Image 2.0 release notes / what's new

xAI has announced the general availability of Imagine Image 2.0, a model designed for professional creative work. The update introduces improved instruction following, high-fidelity typography and layout planning, and a suite of precise editing tools to facilitate iterative design workflows.

High-Fidelity Generation and Typography

Imagine Image 2.0 focuses on producing usable assets for real-world work by closely following detailed instructions. A key technical improvement is the model's ability to plan typography and layout similarly to a human designer, which ensures that dense, multi-part visuals remain cohesive and small text remains sharp.

Precise Editing Capabilities

To support iterative creative processes, Imagine Image 2.0 introduces several specialized editing tools designed to modify specific elements of an image without affecting the rest of the composition:

  • Magic Wand: Edits a specific pointed region while leaving other areas untouched.
  • Segmentation: Allows for the selection of precise areas for modification.
  • Background Removal: Exports subjects with transparent backgrounds.
  • Multi-ref Editing: Supports up to five input images in a single generation to reduce the need for manual compositing.
  • Smart Resize: Enables users to pick a ratio and have the model fill in the frame. Supported ratios include 1:2, 9:16, 2:3, 3:4, 1:1, 4:3, 3:2, 16:9, and 2:1.

Performance Benchmarks

According to xAI, Imagine Image 2.0 was trained for high fidelity across photography, design, and illustration. As of August 7, 2026, the model ranks second globally in both text-to-image generation and image editing based on Elo ratings from the Arena Image Edit and Text-to-Image leaderboards (where xAI models are listed under SpaceXAI).

Workflow Templates and World Building

Creative Templates

xAI has introduced ready-made templates that package common workflows into starting points. These templates cover various tasks, including:

  • Photo Tools: Photo Edit, Reimagine, Photo Collage, Professional Headshot, and BG Removal & Change.
  • Product/Marketing: Product Color Change, Editorial Product Poster, E-Commerce Photos, UGC Photos, and Merch Maker.
  • Design Tools: Mascot Maker, Icon Maker, and Character Sprite.
  • Game Assets: Props & UI Kit.
  • Streaming: Emoji Creator.

World Building for Video

The model supports the creation of consistent visual worlds. Users can generate characters, locations, and props separately while maintaining a consistent style across different images, a capability intended to assist in the development of assets for video production.

Availability and API Access

Imagine Image 2.0 is available as the "Quality Mode" on grok.com/imagine and within the Grok iOS and Android applications. For developers, the model is accessible via the API as grok-imagine-image-2.0.

Sources

Related

  • Dispatch
    Grok Imagine Quality Mode API Release

    xAI has launched Quality Mode for the Grok Imagine API, providing enterprise developers with enhanced realism, improved multilingual text rendering, and superior creative control for image generation and editing.

  • Dispatch
    xAI Grok Imagine API Release

    xAI has launched the Grok Imagine API, a unified video-audio generative model designed for high-quality video generation and precise video editing with a focus on low latency and cost-efficiency.

  • Dispatch
    ChatGPT Images 2.0 Release

    OpenAI has released ChatGPT Images 2.0, a significant update to its image generation capabilities featuring improved precision, multilingual text rendering, and enhanced stylistic realism.

  • Dispatch
    Grok Imagine Video 1.5 Release Notes

    xAI has released Grok Imagine Video 1.5, an image-to-video model featuring improved motion, physics, and integrated audio generation with a faster generation speed.

  • Dispatch
    Imagine Video 1.5 with References release notes

    xAI has updated Imagine Video 1.5 to include text-to-video generation, native 1080p resolution, and image and voice reference capabilities for consistent character and scene generation.