Image · xai/grok-imagine-image-2
Grok Imagine Image 2.0 — the new generation of xAI images
The fresh flagship of the Imagine line: one model both generates from scratch and edits an uploaded frame from a text instruction. It adds a quality knob, raises output to 2K on the long edge and offers over a dozen aspect ratios. Handles detailed prompts and real-world entities — brands, places, recognizable objects.
from 6,6 ₽ per image
When to pick Image 2.0
- You need a file up to 2K — for print, covers or marketplaces.
- Editing an existing frame: "make the sky a sunset" instead of rebuilding the prompt.
- Prompts with brands, places and real objects — the model knows the world.
What is new versus the first Grok Imagine
The headline is editing: upload an image, describe the change, and the model applies it while keeping the structure of the original. Second is the quality control: "medium" for maximum fidelity, "low" for fast cheap drafts. Plus honest 2K output and a much wider list of aspect ratios.
How to write prompts
Detailed prompts beat short ones: subject, lighting, mood, style directives at the end. Name real things directly — the model handles brands and places well. When editing, describe only the change, not the whole scene. The credit price is shown before you run.
Image 2.0 FAQ
How is it different from the first Grok Imagine Image?
It edits uploaded images, outputs up to 2K and adds a low/medium quality control. The first version remains the fastest and cheapest of the line.
Can I upload my own photo?
Yes — attach a frame and describe the change in text. The model applies the edit while preserving the original composition. It takes one input image.
Which quality should I choose?
"Medium" for final images — that is the fidelity ceiling. "Low" for quick drafts and idea exploration: cheaper and noticeably faster.
What does a generation cost?
Charged in credits; the amount depends on the settings and is shown before the run and on the pricing page.