Skip to main content

Head-to-head comparison

GPT Image 2 vs. MAI-Image-2.5

Compare arena scores with documented retrieval dates, task-specific image pricing, and verified specifications. Text-to-image generation and image editing are never blended into an artificial overall score.

11

shared result series

4

shared sample prompts

2026-08-16

latest arena retrieval

Change image model selection

Specifications and cost

General specifications and task-specific pricing

A model can have one price for generating new images and another for editing. Prices therefore remain tied to the corresponding arena configuration.

Specifications and pricing for GPT Image 2 and MAI-Image-2.5
AttributeGPT Image 2MAI-Image-2.5
ProviderOpenAIMicrosoft AI
ReleasedApril 2026June 2026
AccessAPI, Web appAPI, Web app
LifecycleNo separate lifecycle noticePreview in Microsoft Foundry Status source
WeightsProprietaryProprietary
ParametersNot published20 billion
Maximum output4K1,048,576 pixels
Arena configurationHighNo separate configuration named
Text-to-image price$211 per 1,000 images$48.1 per 1,000 images
Editing price$211 per 1,000 images$48.1 per 1,000 images
Technical sourceModel sourceVerified 2026-08-15Model sourceVerified 2026-08-15
GPT Image 2MAI-Image-2.5
Text to Image ArenaUSD per 1,000 images at 1,024 × 1,024 pixels
GPT Image 2$211
MAI-Image-2.5$48.1Lower price
Image Editing ArenaUSD per 1,000 images at 1,024 × 1,024 pixels
GPT Image 2$211
MAI-Image-2.5$48.1Lower price

Independent preference tests

Shared arena benchmarks

Artificial Analysis and Arena derive scores from blind comparisons. Raters see outputs for the same prompt and choose the image they prefer. Separate series cover text rendering, photorealism, portraits, art, commercial design, and image editing. Confidence intervals indicate how certain a measured lead is.

11 shared result series
GPT Image 2MAI-Image-2.5
Text to Image ArenaBlind preference from the same prompt, higher is better
GPT Image 21,370 EloHigher arena score
MAI-Image-2.51,310 Elo
Image Editing ArenaBlind preference from the same prompt, higher is better
GPT Image 21,258 EloNo clear arena lead
MAI-Image-2.51,257 Elo
Arena Text to Image: 3D modelingBlind preference from the same prompt, higher is better
GPT Image 21,360.375 arena pointsHigher arena score
MAI-Image-2.51,246.107 arena points
Arena Text to Image: ArtBlind preference from the same prompt, higher is better
GPT Image 21,370.438 arena pointsHigher arena score
MAI-Image-2.51,277.636 arena points
Arena Text to Image: CartoonBlind preference from the same prompt, higher is better
GPT Image 21,396.031 arena pointsHigher arena score
MAI-Image-2.51,270.066 arena points
Arena Text to Image: Commercial designBlind preference from the same prompt, higher is better
GPT Image 21,389.556 arena pointsHigher arena score
MAI-Image-2.51,260.487 arena points
Arena Text to Image: OverallBlind preference from the same prompt, higher is better
GPT Image 21,380.469 arena pointsHigher arena score
MAI-Image-2.51,256.13 arena points
Arena Text to Image: PhotorealismBlind preference from the same prompt, higher is better
GPT Image 21,378.291 arena pointsHigher arena score
MAI-Image-2.51,253.013 arena points
Arena Text to Image: PortraitsBlind preference from the same prompt, higher is better
GPT Image 21,428.866 arena pointsHigher arena score
MAI-Image-2.51,259.199 arena points
Arena Text to Image: Text renderingBlind preference from the same prompt, higher is better
GPT Image 21,424.046 arena pointsHigher arena score
MAI-Image-2.51,282.791 arena points
Arena Image Editing: OverallBlind preference from the same prompt, higher is better
GPT Image 21,462.727 arena pointsHigher arena score
MAI-Image-2.51,401.879 arena points
Arena benchmark scores for GPT Image 2 and MAI-Image-2.5
BenchmarkGPT Image 2MAI-Image-2.5Source
Text to Image Arena
1,370 EloHigher arena score
Rank 1, 95% confidence interval: ±10, 13,336 ratings
1,310 Elo
Rank 5, 95% confidence interval: ±9, 14,529 ratings
Artificial AnalysisRetrieved 2026-08-15
Image Editing Arena
1,258 EloNo clear arena lead
Rank 2, 95% confidence interval: ±6, 18,142 ratings
1,257 Elo
Rank 3, 95% confidence interval: ±8, 11,229 ratings
Artificial AnalysisRetrieved 2026-08-15
Arena Text to Image: 3D modeling
1,360.375 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±9.77, 6,635 ratings
1,246.107 arena points
Rank 5, 95% confidence interval: ±9.6, 4,252 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Art
1,370.438 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±8.48, 8,762 ratings
1,277.636 arena points
Rank 3, 95% confidence interval: ±8.37, 6,571 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Cartoon
1,396.031 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±6.88, 28,210 ratings
1,270.066 arena points
Rank 7, 95% confidence interval: ±6.37, 17,440 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Commercial design
1,389.556 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±6.61, 27,138 ratings
1,260.487 arena points
Rank 6, 95% confidence interval: ±6.3, 17,261 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Overall
1,380.469 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±4.62, 69,194 ratings
1,256.13 arena points
Rank 8, 95% confidence interval: ±4.57, 44,940 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Photorealism
1,378.291 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±6.62, 26,719 ratings
1,253.013 arena points
Rank 10, 95% confidence interval: ±6.3, 17,468 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Portraits
1,428.866 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±8.77, 13,521 ratings
1,259.199 arena points
Rank 10, 95% confidence interval: ±7.61, 8,541 ratings
ArenaRetrieved 2026-08-16
Arena Text to Image: Text rendering
1,424.046 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±7.09, 26,226 ratings
1,282.791 arena points
Rank 7, 95% confidence interval: ±6.6, 15,799 ratings
ArenaRetrieved 2026-08-16
Arena Image Editing: Overall
1,462.727 arena pointsHigher arena score
Rank 1, 95% confidence interval: ±3.94, 184,189 ratings
1,401.879 arena points
Rank 3, 95% confidence interval: ±3.85, 155,115 ratings
ArenaRetrieved 2026-08-16

Visual comparison

Sample images from documented runs

Our images complement preference data, but do not replace it. The prompt, request count, and selection rule match. Provider resolution, quality tier, randomness, and prompt processing remain visible limitations, so this is not a fully controlled laboratory test.

Largely standardized comparison protocol. Both models receive the same archived prompt, exactly one request, and the first result counts. Files are normalized to 1,024 × 1,024 pixels without cropping. Provider resolution, randomness, quality tier, and prompt processing can still differ. These images are a practical visual test, not a reconstruction of the arena configuration.

Character and detail

Anatomy, clothing, and spatial consistency

GPT Image 2, Character and detail
GPT Image 2
MAI-Image-2.5, Character and detail
MAI-Image-2.5
Prompt and settings

Create a square cinematic full-body portrait of a fictional bicycle courier waiting under a transparent umbrella at a rainy tram stop in Hamburg at blue hour. She wears a mustard raincoat, navy trousers, red sneakers, a silver helmet, and carries a teal messenger bag. Keep both hands visible, render the bicycle correctly, use realistic wet-street reflections, and include no readable brand names.

GPT Image 2. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-2, generated Aug 14, 2026, 12:13 PM. Source
MAI-Image-2.5. Provider OpenRouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048508, model ID microsoft/mai-image-2.5, generated Aug 14, 2026, 1:26 PM. Source

Protocol gradually-image-comparison-v1, selection rule: first result.

Prompt SHA 256: 985a75b38642a9f0e5e0c0e0c67f1b7f67780d42bd2168fa24e9a7a7a94b0da6

GPT Image 2, image SHA 256: 82f6e2369ab4fb6efb0ef48feeff0d14f1d6087c64c35113a983456f0f0879ac

GPT Image 2, request SHA 256: db4ca97dc43effd578430b77f8ca559c7dd5cbe60e663ba9ecc2e89a33480b0d

MAI-Image-2.5, image SHA 256: ca1c58688ae4601975d7e1f7b58bb43a0a999e736f610d9f51cd11e06284bced

MAI-Image-2.5, request SHA 256: 40fc54cbcee64f9045c44e702e013b649a0d8bbd7f8e2390d839bc38ee183b90

Infographic

Numbers, labels, and visual organization

GPT Image 2, Infographic
GPT Image 2
MAI-Image-2.5, Infographic
MAI-Image-2.5
Prompt and settings

Create a square German infographic titled SO FUNKTIONIERT PHOTOSYNTHESE. Show exactly four numbered steps in this order: 1 LICHT, 2 WASSER, 3 CO₂, 4 ZUCKER + SAUERSTOFF. Use a clean editorial science style with one plant cross-section, simple arrows, high contrast, legible labels, and no other text.

GPT Image 2. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-2, generated Aug 14, 2026, 12:14 PM. Source
MAI-Image-2.5. Provider OpenRouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048538, model ID microsoft/mai-image-2.5, generated Aug 14, 2026, 1:26 PM. Source

Protocol gradually-image-comparison-v1, selection rule: first result.

Prompt SHA 256: ed65cb4e6a51e2336f8b149ad4cbb16e020f07d527ad94362e719d2dcf91d94f

GPT Image 2, image SHA 256: 6c15466e258e2b14a8e862959cb6a67b4f7a23896ea5d42c5a5618d848847711

GPT Image 2, request SHA 256: ed741d3d120188b9286d3576e7dcd3ace9dd7792978d4d5426022f32ed7ca381

MAI-Image-2.5, image SHA 256: eaf00a906231e8859f98ef5dc2cce7e8017797819aea453305a4ef0e970d4dbf

MAI-Image-2.5, request SHA 256: bbedc701e9933a8d6ed0fbb2a6fbb7214214fad9d32d9ae80a114a7254a0bcf1

Product photography

Materials, reflections, and fine details

GPT Image 2, Product photography
GPT Image 2
MAI-Image-2.5, Product photography
MAI-Image-2.5
Prompt and settings

Create a square premium product photograph of a brushed titanium wristwatch standing upright on dark green marble. The watch has a cream dial, thin black hands set to 10:10, twelve distinct hour markers, a realistic crown, and a dark brown leather strap. Soft window light from the left, controlled reflections, shallow depth of field, no text, no logo, no extra objects.

GPT Image 2. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-2, generated Aug 14, 2026, 12:13 PM. Source
MAI-Image-2.5. Provider OpenRouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048513, model ID microsoft/mai-image-2.5, generated Aug 14, 2026, 1:25 PM. Source

Protocol gradually-image-comparison-v1, selection rule: first result.

Prompt SHA 256: c67fd71e91304458a2feef5c91efa8677260fab0c0099a29a2c52cb579f5b2f3

GPT Image 2, image SHA 256: 471c71b30618afb1a8a67ab8363308d67f99c65ef6740ca72417806d06d3c346

GPT Image 2, request SHA 256: 7b27ff2c9b64529a36bebf2d572c3448f26e68f3e9d1ac6a3f2b5ead8698f60f

MAI-Image-2.5, image SHA 256: 67d4930b711fac61482231c4bc270cfcdb41c1b0daf653045781fb2126e237dd

MAI-Image-2.5, request SHA 256: 071e156eeb3e65e6744b7c6b675ffcb32b551dcfc1033a55e30e9561c0bd78a9

Typography and layout

Legible text, hierarchy, and composition

GPT Image 2, Typography and layout
GPT Image 2
MAI-Image-2.5, Typography and layout
MAI-Image-2.5
Prompt and settings

Create a square editorial poster for an imaginary night train called MONDFALTER. Show the exact German headline MONDFALTER and the exact subline BERLIN NACH LISSABON. Use a restrained midnight-blue and warm-cream palette, one stylized moth, strong Swiss-grid typography, generous negative space, and no additional words or logos.

GPT Image 2. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-2, generated Aug 14, 2026, 1:59 PM. Source
MAI-Image-2.5. Provider OpenRouter, aspect ratio 1:1, provider default 1K resolution, n=1, seed=provider random, actual cost $0.048473, model ID microsoft/mai-image-2.5, generated Aug 14, 2026, 2:03 PM. Source

Protocol gradually-image-comparison-v1, selection rule: first result.

Prompt SHA 256: 2e012c3d14ee5dd02203a94f394ca80a5038cdebd6032b25a8b8a221f2255419

GPT Image 2, image SHA 256: 434cf9939b9be2a2c2711e13d027c9438c47038bb32229ceeaf405d2e89d2afc

GPT Image 2, request SHA 256: 473a774dfe9f9b7bb597e9e8b63af4e770b6bc4e7f8bc90682636db3516e7353

MAI-Image-2.5, image SHA 256: 4c8a60555b57b8c3b5ebaa131371d1bba9d31bfafe7aaa772554969f7c28cb08

MAI-Image-2.5, request SHA 256: f14577273bd6b0980f49e13988b01555bb55dc824317b201b5fab3d2ef72e666

How to read this comparison

An Elo score is not a verdict on creativity

A higher arena score means this specific configuration was preferred more often in blind comparisons.

Typography, character consistency, local execution, licensing, and price may matter more to your workload than overall visual appeal.

Prices cover 1,000 images at 1,024 × 1,024 pixels using settings documented by Artificial Analysis. Subscriptions and special rates are excluded.

More matchups

Related image model comparisons

Compare GPT Image 2 and MAI-Image-2.5 with other important image models using the same benchmark and sample methodology.

View the complete image model comparison