1
Choose Generate or Edit
Start from a text prompt for a brand-new image, or upload 1–5 reference images to edit and transform an existing visual.
Grok ImagineGrok ImaginePrecise AI image generation and editing, built for real creative work
Generate new images from text or edit existing ones with up to 5 reference images. Grok Imagine Image 2.0 follows instructions down to the details, plans typography and layout like a designer, and keeps even small text sharp — so posters, infographics, product shots, and illustrations come out usable, not just pretty.
Create with Grok Imagine Image 2.0 in 4 Simple Steps
Start from a text prompt for a brand-new image, or upload 1–5 reference images to edit and transform an existing visual.
Describe the scene, style, layout, and any exact text you want rendered. Image 2.0 rewards specific prompts with precise results.
Select 1:1, 3:2, 2:3, 16:9, or 9:16 depending on where the image will be used.
Submit the task, watch the result appear in the gallery, then download the finished image or refine it with another edit pass.
Designer-Grade Generation and First-Class Editing
Image 2.0 plans typography and layout the way a designer would, so dense multi-part visuals like posters, infographics, and covers hold together — and small print comes out sharp and readable.
The model follows your prompt down to the details. Complex scenes, multiple objects, specific art styles, and exact wording land the way you describe them, cutting down on retry loops.
Switch to edit mode by uploading 1–5 images. Describe the change you want and Image 2.0 alters exactly that while preserving the original subject, style, and composition.
Image 2.0 preserves what you put in across generations and edits, so you can build a character, their locations, and props as separate images that hold one consistent look.
Trained for fidelity across photorealistic photography, graphic design, and illustration — from cinematic portraits to pixel art. At launch it ranked second in the world in both text-to-image and image editing on the Arena leaderboards.
Generate in 1:1, 3:2, 2:3, 16:9, or 9:16 to fit social posts, thumbnails, stories, posters, and print layouts without cropping after the fact.
How to Generate and Edit Images with Grok Imagine 2.0
This page integrates xAI's Grok Imagine Image 2.0 model directly in the browser: text-to-image generation plus reference-based image editing, with the model already selected in the playground above.
Use the playground at the top of this page. The Image 2.0 model is preselected, so there is nothing to configure before your first generation.
Leave the input empty to generate from text alone, or upload up to 5 images to switch into edit mode — one image for a targeted edit, several to combine subjects, styles, or products in a single result.
Image 2.0 is built for instruction following. Spell out the composition, style, mood, and any exact wording (put it in quotes). For edits, state exactly what should change and what must stay the same.
Choose 1:1 for square posts, 16:9 for thumbnails and banners, 9:16 for stories and shorts, or 3:2 / 2:3 for photo and poster formats.
Check the credit cost shown on the button, submit, and preview the result. Download what you like, or iterate with another prompt — failed generations are refunded automatically.
Image 2.0 renders typography unusually well. Write the precise headline or label you want — like "SUMMER SALE — 50% OFF" — and it will appear sharp, even at small sizes.
In edit mode, name the one thing to change and explicitly keep the rest: 'replace the red jacket with a navy coat, keep the face, pose, and background unchanged.'
Upload the same character or product photo as a reference across multiple generations to keep identity, colors, and style consistent through a whole set of images.
For posters and infographics, specify the structure — 'bold stacked headline on the left, hero photo upper right, three feature blocks along the bottom' — and Image 2.0 will compose it cleanly.
Precise Image Generation for Every Workflow
Produce scroll-stopping square, vertical, and widescreen visuals with real, readable text overlays — no separate design tool needed.
Create posters, ad creatives, and product shots with accurate branding text, then edit colors, backgrounds, or copy without regenerating from scratch.
Explore layouts, mockups, icons, game assets, and illustration styles fast, keeping one visual language across a whole series of images.
Turn dense information into clean infographics, diagrams, tutorial sheets, and covers where labels and small print stay legible.
Everything About the Grok Imagine 2.0 Image Generator
Grok Imagine Image 2.0 is xAI's latest image generation and editing model, released in August 2026 and built for precise, production-ready visuals. This page integrates it directly, so you can generate and edit images in the browser without an xAI API key.
Image 2.0 follows instructions much more closely, plans typography and layout like a designer, keeps small text sharp, and treats editing as a first-class capability with strong identity preservation. At launch it ranked second worldwide in both text-to-image and image editing on the Arena leaderboards.
Yes. Upload 1–5 images and describe the change — swap a background, change a product color, restyle a portrait, or replace text on a poster. The model changes what you ask for and preserves the rest of the image.
You can upload up to 5 images in a single generation. Use one for a targeted edit, or several to combine a character, product, and style into one composition without manual compositing.
This integration supports 1:1, 3:2, 2:3, 16:9, and 9:16, covering square posts, photo formats, posters, thumbnails, and vertical stories.
Yes — typography is its standout strength. It plans layouts the way a designer would, so posters, infographics, and covers with headlines, labels, and even small print come out coherent and readable. Put the exact wording you want in quotes in your prompt.
Both text-to-image generation and image editing cost 5 credits per image — the same rate as our standard Grok Imagine image model. If a generation fails, the credits are refunded automatically.
New accounts receive free welcome credits, and you can earn more through daily check-in rewards — enough to try several generations before deciding on a plan. The credit cost is always shown before you submit.
Commercial use depends on the current model, platform, and account terms that apply to you. Review the applicable xAI usage policies and your plan's terms, and ensure your prompts and references don't infringe third-party rights.
No. This page runs Grok Imagine Image 2.0 through our own integration, so you only need an account here with credits — no X (Twitter) subscription, xAI API key, or extra setup.