Mixer AI
Grok Imagine
Create images from a description and choose the version around whether you need to edit an uploaded photo.
When to choose it
Grok Imagine has two distinct workflows. Version 1.0 creates from text and edits one uploaded photograph. Version 2.0 starts from text and offers Segment Edit for eligible generated results. Decide whether you are starting with your own source image or a new idea before selecting a version.
- Text-based creation for illustrations, product scenes and visual drafts.
- Separate editing paths: an uploaded photo in 1.0 or an eligible 2.0 result through Segment Edit.
Modes and settings
- 1.0: text with Speed or Pro priority, or editing one photo.
- 2.0: new images from text; Segment Edit starts from an eligible generated result. Five aspect ratios are available for generation.
Price in coins
| Version | Quality / size | Input | Coins / generation |
|---|---|---|---|
| Grok Imagine 1.0 | — | Photo editing | 5 |
| Grok Imagine 1.0 | — | Text | 5 |
| Grok Imagine 1.0 | pro | Text | 5 |
| Grok Imagine 2.0 | — | Segment Edit | 5 |
| Grok Imagine 2.0 | — | Text | 5 |
Coins are Mixer AI balance units. The table shows a base request without extra variations. Multiple results and new attempts can increase the total; check it before submitting.
What you can make
- Explore a visual idea for a post.
- Change a detail in an eligible version 2.0 result.
How to direct the result
- For new images, describe the entire scene: subject, place, composition, light and medium.
- For edits, identify one detail and its replacement rather than changing the whole concept.
Example: a social-post illustration
A small red boat beside a wooden jetty on a calm lake. Early morning, thin mist and soft reflections. Place the boat in the lower-right third with plenty of open space above. Colored-pencil illustration on lightly textured paper, no lettering or border.
What to keep in mind
- Segment Edit is not ordinary uploading of an arbitrary photograph into version 2.0.
- Inspect lettering, logos and small objects separately.
Sources
- Prices and availability — Mixer AI catalog, updated 2026-09-18.
- Official maker site: xAI
Model facts verified: 2026-09-20.
FAQ
Which version accepts my photo?
Use 1.0 for ordinary uploaded-photo editing. It accepts one image.
How does Segment Edit work?
It continues work on an eligible version 2.0 result. Select the result and describe the change.
Are Speed and Pro available everywhere?
No. They are priorities for 1.0 text generation, not universal switches for every version or edit.
Can I build a scene from ten photos?
This page does not support that input. Use another model for a set of visual references.
Can text exactly reproduce an existing photograph?
Text defines a new scene rather than copying every source detail. Use a photo-input workflow when a specific original matters.
What if a local edit fails?
Keep the previous result, narrow the requested change and try again from the successful base.
What changes the price?
It depends on version, quality and mode. The table prices one generation at the listed settings. Additional results, new attempts and chargeable input materials can increase the total.
Do I need a separate model subscription?
No. In Mixer AI, you top up one coin balance and use it for the tasks you need. You do not need a separate subscription to this model.