Grok Imagine Quality

What it does

Grok Imagine Quality is xAI's image editing model, described as producing high-fidelity image edits at its quality=hd setting. It is image-to-image only: you supply a source image and an instruction, and it returns an edited image. Its listed capability does not include creating pictures from a text prompt alone. Through the Image Editor pipeline on pipe2 you choose the output aspect ratio from 1:1, 16:9, 9:16, 4:3 and 3:4, so the edit and the reframe happen in one run.

When to use it

• Running the Image Editor pipeline to restyle an existing image, changing the overall look while keeping the underlying subject.
• Removing objects from a photo you already have, rather than regenerating the scene from scratch.
• Modifying specific elements of an image (a garment, a background detail, a colour) with a written instruction.
• Transforming an image end to end, when the source is a starting point rather than something to preserve.
• Producing the same edit in a different frame, using the pipeline's 1:1, 16:9, 9:16, 4:3 and 3:4 aspect ratios for different placements.

What to know before you run it

A source image is required. Its only listed capability is image-to-image, so a run without an input image has nothing to edit. It is described at the quality=hd setting, and no other fidelity setting is documented for this model here; if you want a different trade-off, the Image Editor's model selector is where you pick among the models available for that pipeline. Billing is metered in a single unit, one Grok Imagine Quality image edit, and the live price table on this page is authoritative for what a run costs. Three worked examples are published on pipe2 if you want to see the inputs before spending credits.

Also known as

You may see this listed as Grok Imagine Image Quality or Grok Imagine Quality Mode; both refer to the same model, whose slug is grok-imagine-image-quality.