Elon Musk’s AI company xAI has introduced a new version of its image-generation tool, Grok Imagine, called “Imagine Image 2.0.”
The new version offers improved capabilities not only for generating images but also for editing them more accurately and adding clear, legible text within images. According to the company, the new model is focused on creating images for real-world and professional use.
According to xAI, “Imagine Image 2.0” is now available as a Quality Mode on the web version of Grok, while it can also be used through the iOS and Android apps. The company has currently made it available to general users.
Grok announced through its official account that “Imagine Image 2.0” is its next-generation image model, featuring more precise editing, clearer text rendering, improved factual accuracy, and greater usefulness for real-world tasks.
The company says the model is not intended merely to create beautiful images, but also to assist with tasks where accurate modifications and clear details are essential.
One of the key features of the new model is its editing tools. According to Grok, the Magic Wand allows users to modify only a specific part of an image while leaving the rest of the image unchanged.
Similarly, segmentation allows users to select a particular area of an image more precisely.
The Smart Resize feature enables users to easily change an image’s framing and dimensions according to their needs.
These features could be particularly useful for graphic designers, social media content creators, and people working in marketing and advertising, as they often need to modify only a specific part of an image rather than recreate the entire image from scratch.
Announcing Imagine Image 2.0, our next generation image model with precision editing, crisp text rendering, improved factuality, and real world usefulness.
Image 2.0 helps you make images for real work.https://t.co/y423lvT7mL
— Grok (@grok) August 8, 2026
“Imagine Image 2.0” also improves the model’s ability to generate text within images. AI-generated images often suffer from problems involving incorrectly rendered words, logos, signboards, and complex typography. xAI says the new model offers clearer and more accurate text rendering, making it easier to create designs in which text is an important part of the image.
The company has also claimed improvements in the model’s “factuality,” meaning its ability to better understand real-world objects and information.
However, this is a claim made by the company itself and does not mean that every image generated by the AI will necessarily represent reality accurately.
Information released by Grok also states that “Imagine Image 2.0” has achieved strong positions in various competitive evaluations of image generation and image editing.
Grok’s official account claimed that the model ranks second in the world on Arena rankings in both text-to-image generation and image editing. However, to independently assess this ranking, it is important to consider the specific version of Arena, the time of testing, and which models were included in the comparison.
Following the launch of the new model, users also shared their reactions on social media. Some users described its new capabilities as a major improvement, particularly praising its enhanced typography, complex designs, and ability to edit specific parts of an image.
However, not all users were positive. Some said that the results had become worse compared with the previous version and that the images felt more artificial. This shows that AI image models do not necessarily produce the same results for every user or every type of prompt.
Overall, the arrival of “Imagine Image 2.0” has intensified competition in the AI image-generation space. The model’s main significance lies in its greater focus on practical tasks such as modifying specific parts of an image, resizing images, and adding better-quality text, rather than simply generating an image from a single prompt.
xAI says that “Imagine Image 2.0” was built for real creative work. It remains to be seen how accurate these claims prove to be in the hands of general users and professional designers, and where its performance ultimately stands compared with other leading AI image models in the coming days.
