1
Write or Upload
Describe the image you want, or upload up to 5 reference photos to guide an image-to-image generation.
Grok ImagineGrok ImagineCreate images from a prompt or transform your reference pictures
Use Grok Imagine for text-to-image generation, image-to-image editing, and fast visual exploration. Write a detailed prompt or upload up to five reference images, then create social graphics, thumbnails, product concepts, characters, and artwork in the aspect ratio you need.
Generate AI Images in 4 Simple Steps
Describe the image you want, or upload up to 5 reference photos to guide an image-to-image generation.
For text-to-image, pick 1:1, 16:9, 9:16, 3:2, or 2:3. Toggle Pro Mode when you want a more detailed generation.
Review the estimated credit cost, submit the generation, and let Grok Imagine produce a new image from your prompt and references.
Inspect the result at full size, refine the prompt or references when needed, then download the image for your project.
Prompt-Based Creation, Reference Editing, and Flexible Formats
Turn an idea into a visual draft, refine one detail at a time, and generate new variations without rebuilding your workflow. Grok Imagine is well suited to rapid concept exploration.
Describe the subject, setting, composition, lighting, lens, palette, and mood in plain language. More concrete visual direction gives the model a clearer target.
Start from a text prompt or upload up to five reference images. Use image-to-image mode to guide the subject, composition, color, or style of a new result.
Restyle an existing image, explore alternate compositions, or create visual variations from the same references. Always review small text, logos, hands, and product details before publishing.
Choose 1:1, 16:9, 9:16, 3:2, or 2:3 for square posts, vertical stories, landscape thumbnails, banners, portraits, and other common layouts.
Turn on Pro Mode when you want the model to spend more generation credits on a more detailed result. The generator shows the estimated credit cost before you submit.
How to Use Grok Imagine to Create AI Images
Grok Imagine combines text-to-image and image-to-image generation in one browser workflow. Follow these steps to move from an idea or reference image to a finished visual you can review and refine.
Open the playground at the top of this page. Nothing needs to be installed; sign in when you are ready to submit a generation and use your available credits.
For a brand-new visual, stay in text-to-image and describe your idea. To restyle or edit an existing picture, switch to image-to-image and upload up to 5 reference photos to guide the result.
Spell out the subject, environment, visual style, lighting, camera angle, mood, and color palette. If text must appear in the image, include the exact wording and verify it in the result.
Choose the ratio that fits your platform: 1:1 for Instagram, 9:16 for TikTok and Stories, 16:9 for thumbnails and banners, or 3:2 / 2:3 for prints. Turn on Pro Mode when you need higher-fidelity output.
Generate and inspect the output. Tweak the prompt, replace a reference, or change one setting at a time until the result fits your goal, then download it.
Concrete details beat vague adjectives. 'Golden-hour backlight, 85mm portrait, shallow depth of field' gives the model more useful direction than 'beautiful photo'.
Anchor the look with an explicit cue — 'editorial photography', 'Studio Ghibli anime', 'isometric 3D render' — so Grok Imagine knows exactly which direction to take.
Upload a reference photo when a character, product, palette, or composition matters. References guide the output, but important identity and product details should still be checked.
Change one thing at a time — lighting, then framing, then palette. Small, deliberate edits make it easier to understand what improved the result.
Empowering Creators Across Industries
Generate eye-catching thumbnails, social media graphics, and blog illustrations that boost engagement.
Create product images, marketing visuals, and advertising graphics without expensive photo shoots.
Rapidly prototype concepts, generate mood boards, and explore creative directions before finalizing.
Produce classroom demonstrations, textbook illustrations, and research diagrams with visual clarity.
Everything You Need to Know About Grok Imagine
Grok Imagine is an xAI image generation model for creating or transforming visuals with prompts. This page supports text-to-image and image-to-image workflows, including up to five reference images for image-guided results.
You can explore the generator before submitting. Generation requires an account and available credits; eligible new accounts may receive registration credits, while credit packs and plans support continued use.
Common uses include social graphics, thumbnails, concept art, character studies, product concepts, blog illustrations, mood boards, and creative variations of an existing image. Choose the workflow and aspect ratio that fit the final placement.
Use concrete visual details: subject, action, environment, composition, style, lighting, palette, and mood. A focused prompt is usually easier to refine than a long list of conflicting instructions.
Usage rights depend on the applicable model, platform, and account terms, as well as the material in your prompt and reference images. Review the current terms and make sure you have permission to use any brands, people, or source assets before publishing commercially.
Text-to-image supports five aspect ratios: 1:1, 16:9, 9:16, 3:2, and 2:3. Choose the ratio for your intended layout before generating so less cropping is needed later.
Output quality varies with the prompt, references, and selected mode. Pro Mode is available for a more detailed generation and uses more credits. Inspect the downloaded image at its actual size before using it in a final design.
You can request words, signs, or labels by including the exact text in your prompt. AI-generated lettering can still contain spelling or layout errors, so check critical text and add final typography in a design tool when precision matters.
The image generator on this page exposes standard generation and an optional Pro Mode; it does not expose a separate Spicy Mode for images. Normal, Fun, and Spicy are options on the standard Grok video generator.
Write a focused prompt, choose an aspect ratio, add reference images when they provide useful visual guidance, and decide whether Pro Mode is worth the additional credits. After generation, review important details and refine one variable at a time.
Yes. Image-to-image mode accepts up to five reference images. Use your prompt to explain which elements should remain recognizable and what should change, such as the style, setting, lighting, palette, or composition.
You can prompt for photography, editorial illustration, anime-inspired art, watercolor, concept art, pixel art, surreal imagery, or 3D-rendered looks. Describe visual characteristics rather than relying only on a style label.