A still can start from a sentence, or from a photo you want to steer. The free Grok image generator on CanvaPro — labelled Grok Imagine on the page — offers both Text to Image and Image to Image. That second path matters, because you can upload a reference even if older marketing copy on the site once suggested the tool was prompt-only. The live form is the source of truth: it accepts a photo and a prompt.
Image to Image takes a JPG, PNG, or WEBP up to 10 MB. In either mode you write a prompt, choose an aspect ratio, complete the security check, and click Generate Image. A preview typically appears in about a minute so you can judge the result before you download. There is no account to create, and the usual file does not pick up a watermark from the page.
Text to Image and Image to Image
Text to Image invents the frame from your description. Use it when you do not have a starting photo. Image to Image starts from a reference you upload and uses the prompt to guide the edit, restyle, or variation. Only upload a file you have the right to change. A simple, sharp still usually steers the result more reliably than a busy, compressed thumbnail.
Aspect ratios on the page
The form offers Auto, 1:1, 3:2, 2:3, 9:16, 16:9, 3:4, and 4:3. Auto lets the model choose a fit. Pick 1:1 for a square post, 9:16 for a vertical story, 16:9 for a wide hero, and the remaining options for a more editorial still. Set the ratio before you generate so you are not cropping a finished image to fit the placement.
How to generate an image, step by step
- Choose Text to Image or Image to Image. In Image to Image, upload a JPG, PNG, or WEBP that is 10 MB or smaller.
- Write a prompt that names the subject, the setting, the light, and the mood. If you uploaded a reference, say what should change and what should stay.
- Pick an aspect ratio, complete the security check, and click Generate Image once.
- Review the preview. If the composition works, download it. If not, adjust one part of the prompt or swap the reference and run again. Keep earlier downloads so you can compare.
Pick the aspect, write a concrete prompt, and when you retry, go back to the Grok Imagine form with a tighter sentence rather than a stacked download.
Prompts that stay useful
One clear scene is better than a pile of adjectives. Name the place, the time of day, and the main action or object. Avoid packing official logos or tiny legal text into the request if those letters have to be exact. Creators use Grok Imagine for concept stills, landscape moods, and social drafts. Teams use Image to Image to explore a variation of a photo they already own. Treat every result as synthetic media, and check your own rights and labeling needs before a generated picture is used in an ad or a client deck.
If you want another model’s reading of the same brief, the walkthrough to generate images with GPT Image uses a similar text-or-image flow.
If the preview misses the brief
Simplify the prompt, confirm the 10 MB limit in Image to Image, and make sure the security check is still valid. A busy moment can add a short wait. If you would rather compare the same idea on a GPT Image model, open the GPT image generator and run the prompt there. That page uses a different model list and a slightly different aspect set, which can be a useful second pass.
Uploads and prompts travel over HTTPS. Requests are handled to complete generation and are not sold. Only upload reference photos you have the right to use. The live form supports a photo upload, so you can ignore any older note that says not to add one.
A tightly locked edit on a photo you need to keep is often a better fit for Flux Kontext than another loose text-to-image pass.
After the Grok Imagine preview appears
Review the still at the aspect and size you plan to use. If you started from a reference, compare it with the original and check that the parts you wanted to keep — a face, a product shape, a composition — are still intact. If they drifted, say so more plainly in the next prompt, or use a cleaner reference, rather than running the same vague instruction again.
Save the download with a short note of the mode, prompt, and ratio. Because the live form supports Image to Image, you can keep using a photo when you have one; you do not have to switch to a prompt-only workaround. Treat the result as synthetic media, and only upload a reference you have the right to change. When a client or platform needs a label, say that the picture was generated or edited with AI.
Locked edits on a photo you need to keep are a better fit for the guide to generate or edit images with Flux Kontext.
A short checklist
- Pick Text to Image, or Image to Image with a JPG, PNG, or WEBP under 10 MB.
- Write one clear prompt and choose Auto or a specific aspect ratio.
- Click Generate Image once and wait about a minute for the preview.
- Download the still you like, and keep earlier versions for comparison.
When you are ready, choose a mode, describe a single scene, and save the preview only after the composition looks right. One considered still is easier to publish than a long string of near-duplicates.