xAI: Grok Imagine Image 2.0

Grok Imagine Image 2.0 accepts text and image references for generation and revision. With 1K and 2K output choices, it can be used to develop a visual draft and then refine it into a more detailed asset.

Compare models

grok-imagine-image-2.0

AvailablexAIImageSyncNew

Price: From $0.03
Released: Aug 12, 2026
Modalities: Image

About Grok Imagine Image 2.0

Grok Imagine Image 2.0 is the current-generation xAI image model, the one xAI's image guide uses in its examples. It generates from text and edits from reference images, and the xAI guide notes support for combining up to five source images in one edit. It offers 1K and 2K output, so a rough version can be refined into a detailed asset.

Where it works well

Current Imagine generation, used in xAI's own image generation examples.
Edits can draw on several source images, which helps combine subjects or transfer a style onto a scene.
One request can return multiple images, so you can compare candidates side by side.
1K and 2K output choices fit a draft-then-refine workflow.
The same model handles new images and revisions, so you do not switch tools mid-project.

When to choose another model

Exploratory batches add up quickly because each request returns full-detail images; the standard model is lighter for rough drafts.
Aspect ratio is not a selectable option, unlike the first-generation and quality models.
Still images only; for motion use a Grok Imagine video model.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Image editingImage to imageText to imageTokenLab endpoint

POST/v1/images/generations

curl -X POST "https://api.tokenlab.sh/v1/images/generations" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "operation": "text-to-image",
  "quality": "medium",
  "resolution": "1k",
  "model": "grok-imagine-image-2.0",
  "prompt": "A minimalist product photo of matte black headphones on a soft blue background."
}'

Pricing

Official price is the model maker's public baseline. TokenLab price is what you pay for this model on TokenLab.

Image editingOfficial
Image to imageOfficial
Text to imageTokenLab Verified, Official

Pricing

per image

TokenLab price: How pricing works
Official price: How pricing works

	Official priceper image	TokenLab priceper image	Discount
Price	How pricing works	How pricing works	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 3, 2026, 07:25 PM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Grok Imagine Image 2.0 in Console with a prompt ready to edit or send.

Help me create an image with grok-imagine-image-2.0 using text-to-image at /v1/images/generations. Show the result, cost, and any limits I should know.

Use cases

Best for

Text to Image

Style transfer and compositing
Provide a portrait and a reference painting and ask for the portrait in that style, or place a product from one image into another scene.
Iterative art direction
Generate several options at 1K, pick one, and ask for specific changes to lighting or props before producing the 2K final.
Marketing asset production
Create hero images and variations for campaigns from a brief, then edit individual versions without regenerating the whole composition.

Prompt examples

Combine the character in image one with the jacket from image two, standing in the street scene of image three.

Create four poster concepts for a jazz festival, flat colours, bold type area at the bottom.

Take this sketch and render it as a clean 2K product shot on a white background.

FAQ

What is Grok Imagine Image 2.0?

It is xAI's latest Imagine image model. It creates images from text prompts and revises existing images from instructions and reference pictures, with 1K and 2K output choices.

How many reference images can I use?

xAI's image guide says an edit request accepts up to five source images, which you can use to merge subjects, transfer a style or compose a scene. Provide images as public URLs or as encoded data.

How does 2.0 differ from grok-imagine-image-quality?

2.0 is the newer generation and the one xAI uses in its documentation examples; the quality model is a higher-fidelity tier of the earlier line. Compare outputs on your own prompts, since the best choice depends on the subject.

Can I generate several images at once?

Yes. xAI's guide describes returning multiple images from one generation request, which makes it easy to compare candidates before choosing one to edit further, which saves repeated requests.

What resolutions are available?

You can choose 1K or 2K output. A common approach is to explore ideas at 1K and then render the chosen concept at 2K for the final asset.

How much does Grok Imagine Image 2.0 cost?

On TokenLab, Grok Imagine Image 2.0 costs How pricing works . The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Grok Imagine Image 2.0 use?

Use https://api.tokenlab.sh/v1/images/generations for Grok Imagine Image 2.0. The request example below shows the matching code shape.

Which operations does Grok Imagine Image 2.0 support?

Grok Imagine Image 2.0 supports Image editing, Image to image, Text to image. Select an operation above to see its endpoint and request example.