Grok Imagine Image 2.0 renders an image from a description or edits from one to five of your own references, with five shapes in text mode and an automatic setting when you edit.
Both starting points sit on one product. Text to image builds a picture from a description alone; image to image starts from one to five uploaded references and follows your description of the change.
The control set stays short on purpose. You choose a shape, write the prompt and add references where the mode allows them; there is no quality tier, file type or seed to configure, and adding references does not change the quote.
Choose a shape, write the prompt and, in editing mode, upload one to five references. The shape list gains an automatic setting when you edit, and the quote for your exact settings is shown before you confirm.
Combine a base picture with supporting images, for example a subject, a colour reference and a setting, and describe how they should come together.
Editing offers automatic shape selection alongside the five explicit aspect ratios. The model chooses the output shape when automatic is selected.
A description can run to 8000 characters, and each reference can weigh up to 30 MB in JPEG, PNG or WebP.
Start from a photo you already have, keep the subject and describe the scene it should end up in.
Pick 16:9 for a header, 9:16 for a story or 1:1 for a feed card in text mode instead of cropping afterwards.
The 8000-character room is enough for framing, style and detail in one prompt, which suits a model without extra sliders.
Use a button on this page. It opens the right workspace with this model already selected, in the language you are reading now.
Write what should happen in the shot or image. Concrete subjects, framing and action give the model more to work with than short labels.
Choose length or quality tier, shape and sound where offered, then read the credit quote for those exact settings before you generate.
Generation uses account credits. Cost depends on the model, quality and output size or duration. Reference images, video or audio may add charges in some modes. Check the workspace quote for your selected inputs and settings before generating; the pricing page lists credit packages.
See credit packagesBetween one and five per editing task, in JPEG, PNG or WebP and up to 30 MB each. Text generation takes no references.
Generation and editing are quoted the same, and the quote does not rise with the number of references. The exact amount for your settings is shown before you confirm.
Text mode offers 1:1, 2:3, 3:2, 16:9 and 9:16. Editing also offers automatic shape selection.
No. There is no quality tier or file type selector on this model. One paid single-reference square edit we ran returned a 1408 × 1408 JPEG, which is not a promise for other settings.
You can describe a local change, but there is no mask or region selection here. Review the result because other parts of the image may also change.
Other models on this platform that do a similar job, or that pair well with this one.
The FLUX 3 family product available on this platform: text generation and reference editing at 1K, 2K or 4K, with PNG, JPEG and WebP output.
Read the guideQwen Image 3.0 renders an image from a description or edits from one to three of your own references, at the 1K or 2K tier, with PNG or JPEG output.
Read the guideA preview video model that renders clips from 1 to 15 seconds in three quality tiers, from a description or one starting image.
Read the guide