AI image generator
Turn a written prompt into an image, or bring a photo in as a reference or an edit target, choosing the model and the framing you need.
Draft content — not yet reviewed by an editor.
Results
Generated content will appear here
How to run a task
The workspace keeps the input, the settings and the result on one screen so you can compare a second attempt against the first.
- Sign in, then pick a model from those the catalogue lists as available.
- Write the prompt. Add a reference or edit image when the model supports it.
- Set the aspect ratio and the quality options the model exposes.
- Check the quoted credit cost, then generate and download the result.
What you can control
The controls change with the model you choose; the workspace only offers settings the model actually declares.
- Model: every image model that the catalogue marks available and callable for this site.
- Aspect ratio: the ratios the selected model accepts.
- Quality and render quality: shown when the model exposes them.
- Reference images: one or more inputs, up to the model limit.
Modes and what they need
A mode is offered only when the selected model declares it.
| Mode | Input |
|---|---|
| Text to image | a written prompt |
| Image to image | a prompt and a source image |
| Reference to image | a prompt and a reference image |
| Image edit | a prompt and the image to change |
Limits worth knowing
Separate runs do not keep a subject identical, so a character or product shot needs the same reference image each time. Uploads must be an image type the service accepts and within its size limit; the workspace reports the actual limit when a file is rejected.
Questions
Which image models can I choose?
All image models the catalogue returns as available and callable for this site. The list in the workspace is read from the service, so it can change without a new deploy.
Can I reuse the same prompt?
Yes, the prompt stays in the field after a run, and every result carries the prompt that produced it in your history.
Why is my uploaded image rejected?
Either the type is not accepted or the file is larger than the service allows. The message above the upload control names the limit that was hit.
Explore next
AI video generator
Write a shot, animate a still, or drive a clip with a reference video, then read the quoted credit cost before the job starts.
Grok Imagine Image 2.0
The Grok Imagine Image 2.0 model as this site lists it: text to image, image to image, reference to image and image editing, with the cost quoted per task.
From prompt to finished image
A worked walkthrough of one image task: choosing a model, writing the prompt, reading the credit quote and judging the result.