ZMIME
Image model comparison · 2 models · Updated Nov 22, 2026

Grok Imagine Image Quality vs Recraft V4.1 Utility Pro

Newest: Recraft V4.1 Utility Pro · Self-hosting: No confirmed open weights · Most API hosts: Recraft V4.1 Utility Pro. Not enough arena votes yet for a quality score, so this is a spec comparison.

  1. xAI

    Grok Imagine Image Quality

    Released Apr 3, 2026

    • Text to image
    • Image editing
    • Arena1,391
    • From$0.05/image
    • Max size2K
  2. Recraft

    Recraft V4.1 Utility Pro

    Released May 14, 2026

    • Text to image
    • Image editing
    • Arena1,169
    • From—
    • Max size—
  3. Add a model

    Make it a three-way comparison.

01 — At a glance

Where they differ

Quality is the arena score from blind, side-by-side human votes. Price, length and resolution come from each lab’s own listings.

  • NewestRecraft V4.1 Utility ProGrok Imagine Image Quality Apr 3, 2026 · Recraft V4.1 Utility Pro May 14, 2026
  • Widest inputsSame inputsGrok Imagine Image Quality: Text prompts, Reference images · Recraft V4.1 Utility Pro: Text prompts, Reference images
  • Image editingBoth edit imagesTakes a reference image as input: Grok Imagine Image Quality yes · Recraft V4.1 Utility Pro yes
  • Self-hostingNo confirmed open weightsNo downloadable weights are documented here
  • Most API hostsRecraft V4.1 Utility ProGrok Imagine Image Quality: 1 · Recraft V4.1 Utility Pro: 2
02 — Side by side

Every spec in one table

Dashes mean the data is not published.

Grok Imagine Image Quality vs Recraft V4.1 Utility Pro specifications side by side
SpecificationGrok Imagine Image QualityxAIRecraft V4.1 Utility ProRecraft
Quality (blind human votes)
Documented capabilitiesHigher-fidelity Grok Imagine image model for prompt-driven generation, editing, and visual design workflowsNot published
Model parametersNot publishedNot published
Text to image arenaElo-style score · higher is betterNot published1,169 ±11#29 of 87 · 2,562 votes
Image editing arenaElo-style score · higher is better1,391 ±6#15 of 64 · 37,096 votesNot published
Price
Starting price$0.05/image 1KNot published
Top price$0.07/image 2KNot published
Extra charges$0.01 per input imageNot published
How it is billedPer imageNot published
Output
ProducesImagesImages
Resolutions1K · 1.5K · 2KNot published
Highest resolution2KNot published
Largest image2KNot published
Images per requestUp to 10Not published
Text in the replyNot publishedNot published
Aspect ratiosNot publishedNot published
Inputs
Text promptsYesYes
Reference images (editing)YesYes
Max reference images5Not published
Access
WeightsProprietaryProprietary
API model IDNot publishedNot published
API hosts12 (best)
ReleasedApr 3, 2026May 14, 2026
DocumentationNot publishedNot published
Specification sourcesxAI API pricing · xAI Imagine API guideNot published
Specs checkedNov 2, 2026Not published
03 — Questions

Which should you choose?

Which is better, Grok Imagine Image Quality or Recraft V4.1 Utility Pro?

Neither model has enough arena votes for a reliable quality score yet, so this page compares specs only. Run the same three or four prompts through each and judge the results.

Which is cheaper?

Grok Imagine Image Quality costs from $0.05/image (1K), up to $0.07/image (2K). Recraft V4.1 Utility Pro has no published per-image price. Prices are list rates and change often; check the lab before budgeting.

What is the difference between Grok Imagine Image Quality and Recraft V4.1 Utility Pro?

Grok Imagine Image Quality is an image model from xAI (Apr 3, 2026), taking text prompts and reference images and producing images. Recraft V4.1 Utility Pro is an image model from Recraft (May 14, 2026), taking text prompts and reference images and producing images.

Which is newer?

Grok Imagine Image Quality Apr 3, 2026; Recraft V4.1 Utility Pro May 14, 2026.

Which can edit an existing image?

Takes a reference image as input: Grok Imagine Image Quality yes; Recraft V4.1 Utility Pro yes.

Can I run any of these on my own hardware?

No confirmed open weights. No downloadable weights are documented here.