Head-to-head comparison

GPT Image 1.5 vs. Stable Diffusion 3.5 Large

Compare shared capability tests, arena scores with documented retrieval dates, task-specific image pricing, and verified specifications. Research benchmarks, preference arenas, and our n = 1 samples remain separate.

shared result series: 19
shared sample prompts: 4
latest arena retrieval: 2026-08-26

First image modelSecond image model

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 1.5 and Stable Diffusion 3.5 Large
Attribute	GPT Image 1.5	Stable Diffusion 3.5 Large
Provider	OpenAI	Stability AI
Released	December 2025	October 22, 2024
Access	API, Web app	API, local
Lifecycle	Marked as deprecated by OpenAI Status source	No separate lifecycle notice
Weights	Proprietary	Open weights
Parameters	Not published	8.1 billion
Maximum output	1,536 × 1,024 pixels	1 Megapixel
Arena configuration	High	No separate configuration named
Text-to-image price	$133 per 1,000 images	No arena entry
Editing price	$133 per 1,000 images	No arena entry
Technical source	Model sourceVerified 2026-08-15	Model sourceVerified 2026-08-15

Prompt adherence and specialist tests

Shared capability benchmarks

Every matchup includes the same four Gradually sample tasks. For each task, one archived first output from every model is ranked in three differently ordered, model-blind passes. The published 0-100 value normalizes the mean rank. It must still be read as an n = 1 sample per model. This pair also shares 7 results from published research benchmarks.

GPT Image 1.5Stable Diffusion 3.5 Large

Gradually sample test: Typography and layoutMean of three blind ranks across all 28 first outputs, n = 1 each

GPT Image 1.533.3 / 100

Stable Diffusion 3.5 Large0 / 100

Gradually sample test: Product photographyMean of three blind ranks across all 28 first outputs, n = 1 each

GPT Image 1.537 / 100

Stable Diffusion 3.5 Large3.7 / 100

Gradually sample test: Character and detailMean of three blind ranks across all 28 first outputs, n = 1 each

GPT Image 1.585.2 / 100

Stable Diffusion 3.5 Large3.7 / 100

Gradually sample test: InfographicMean of three blind ranks across all 28 first outputs, n = 1 each

GPT Image 1.549.4 / 100

Stable Diffusion 3.5 Large2.5 / 100

Key to the marks:better valuebetter, but not conclusivetie

Capability benchmark scores for GPT Image 1.5 and Stable Diffusion 3.5 Large
Benchmark	GPT Image 1.5	Stable Diffusion 3.5 Large
WISE Verified overall	Higher benchmark score82.5 %	40.4 %
WISE culture	Higher benchmark score89 %	49 %
WISE time	Higher benchmark score69.17 %	40.83 %
WISE space	Higher benchmark score88.33 %	44.17 %
WISE biology	Higher benchmark score80 %	30 %
WISE physics	Higher benchmark score75.83 %	37.5 %
WISE chemistry	Higher benchmark score77.5 %	20.83 %
Gradually sample test: Typography and layout	Higher blind-rank score33.3 / 100 Mean rank: 19 (20 / 18 / 19) Five rubric scores Headline: 1.67 / 2 Subline: 1.67 / 2 Text control: 1.67 / 2 Visual brief: 1.33 / 2 Layout: 1 / 2	0 / 100 Mean rank: 28 (28 / 28 / 28) Five rubric scores Headline: 0.67 / 2 Subline: 0.67 / 2 Text control: 0.67 / 2 Visual brief: 1.33 / 2 Layout: 1 / 2
Gradually sample test: Product photography	Higher blind-rank score37 / 100 Mean rank: 18 (18 / 16 / 20) Five rubric scores Subject: 1.33 / 2 Time: 1 / 2 Watch details: 1.33 / 2 Photography: 1.33 / 2 Exclusions: 1.33 / 2	3.7 / 100 Mean rank: 27 (28 / 28 / 25) Five rubric scores Subject: 1 / 2 Time: 1.33 / 2 Watch details: 1.33 / 2 Photography: 1.33 / 2 Exclusions: 0.67 / 2
Gradually sample test: Character and detail	Higher blind-rank score85.2 / 100 Mean rank: 5 (5 / 8 / 2) Five rubric scores Scene: 1.33 / 2 Wardrobe: 1.33 / 2 Anatomy: 1 / 2 Props: 1.33 / 2 Environment: 1.33 / 2	3.7 / 100 Mean rank: 27 (28 / 27 / 26) Five rubric scores Scene: 1.33 / 2 Wardrobe: 1 / 2 Anatomy: 0.67 / 2 Props: 0.33 / 2 Environment: 1 / 2
Gradually sample test: Infographic	Higher blind-rank score49.4 / 100 Mean rank: 14.67 (13 / 15 / 16) Five rubric scores Title: 1.67 / 2 Sequence: 1.67 / 2 Labels: 1 / 2 Diagram: 1.67 / 2 Clarity: 1 / 2	2.5 / 100 Mean rank: 27.33 (28 / 27 / 27) Five rubric scores Title: 1 / 2 Sequence: 0.67 / 2 Labels: 0.33 / 2 Diagram: 0.67 / 2 Clarity: 0.33 / 2
Key to the marks:better valuebetter, but not conclusivetie

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.

8 shared result series

GPT Image 1.5Stable Diffusion 3.5 Large

LMArena Text to Image: 3D modelingBlind preference from the same prompt, higher is better

GPT Image 1.51,215.437

Stable Diffusion 3.5 Large963.912

LMArena Text to Image: ArtBlind preference from the same prompt, higher is better

GPT Image 1.51,223.911

Stable Diffusion 3.5 Large997.688

LMArena Text to Image: CartoonBlind preference from the same prompt, higher is better

GPT Image 1.51,243.08

Stable Diffusion 3.5 Large945.603

LMArena Text to Image: Commercial designBlind preference from the same prompt, higher is better

GPT Image 1.51,240.908

Stable Diffusion 3.5 Large937.14

LMArena Text to Image: OverallBlind preference from the same prompt, higher is better

GPT Image 1.51,238.658

Stable Diffusion 3.5 Large938.302

LMArena Text to Image: PhotorealismBlind preference from the same prompt, higher is better

GPT Image 1.51,248.785

Stable Diffusion 3.5 Large928.086

LMArena Text to Image: PortraitsBlind preference from the same prompt, higher is better

GPT Image 1.51,261.416

Stable Diffusion 3.5 Large941.134

LMArena Text to Image: Text renderingBlind preference from the same prompt, higher is better

GPT Image 1.51,254.593

Stable Diffusion 3.5 Large921.402

Key to the marks:better valuebetter, but not conclusivetie

Arena benchmark scores for GPT Image 1.5 and Stable Diffusion 3.5 Large
Benchmark	GPT Image 1.5	Stable Diffusion 3.5 Large	Source
LMArena Text to Image: 3D modeling	Higher score1,215.437 Rank 11, 95% confidence interval: ±6.83, 14,763 ratings	963.912 Rank 73, 95% confidence interval: ±15.9, 1,629 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Art	Higher score1,223.911 Rank 11, 95% confidence interval: ±6.08, 19,937 ratings	997.688 Rank 70, 95% confidence interval: ±9.46, 5,340 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Cartoon	Higher score1,243.08 Rank 10, 95% confidence interval: ±4.87, 55,661 ratings	945.603 Rank 73, 95% confidence interval: ±6.34, 12,274 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Commercial design	Higher score1,240.908 Rank 10, 95% confidence interval: ±4.72, 55,926 ratings	937.14 Rank 74, 95% confidence interval: ±11.06, 3,051 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Overall	Higher score1,238.658 Rank 11, 95% confidence interval: ±3.36, 142,427 ratings	938.302 Rank 74, 95% confidence interval: ±4.5, 23,396 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Photorealism	Higher score1,248.785 Rank 11, 95% confidence interval: ±4.83, 57,631 ratings	928.086 Rank 74, 95% confidence interval: ±7.07, 9,405 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Portraits	Higher score1,261.416 Rank 9, 95% confidence interval: ±5.95, 28,153 ratings	941.134 Rank 73, 95% confidence interval: ±8.89, 5,704 ratings	LMArenaRetrieved 2026-08-26
LMArena Text to Image: Text rendering	Higher score1,254.593 Rank 10, 95% confidence interval: ±4.98, 51,503 ratings	921.402 Rank 74, 95% confidence interval: ±10.85, 3,222 ratings	LMArenaRetrieved 2026-08-26
Key to the marks:better valuebetter, but not conclusivetie

Visual comparison

Sample images

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

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 1.5. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-1.5, generated Aug 14, 2026, 12:17 PM. Source
Stable Diffusion 3.5 Large. Provider fal.ai, requested 1024x1024, n=1, seed=7087247023770270000, estimated cost $0.068157, model ID fal-ai/stable-diffusion-v35-large, generated Aug 14, 2026, 2:37 PM. Source

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

Prompt SHA 256: 985a75b38642a9f0e5e0c0e0c67f1b7f67780d42bd2168fa24e9a7a7a94b0da6

GPT Image 1.5, image SHA 256: 0dff52ccd39400ac66ac8870c9ea937512ff1474688e70118688f420c38c2296

GPT Image 1.5, request SHA 256: 5199631ac6ff07633dd2fb6ab61c918c26e1e02f37e6131828072233ac04527d

Stable Diffusion 3.5 Large, image SHA 256: e748006fe56e442ad347e3f994f18b86ee1739f20dcee063aa7a03ee710c5d54

Stable Diffusion 3.5 Large, request SHA 256: 5d005ee24926d68b7d2299b8ecbe4e14919f652761f6de2568a043f2a6258da0

Infographic

Numbers, labels, and visual organization

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 1.5. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-1.5, generated Aug 14, 2026, 12:17 PM. Source
Stable Diffusion 3.5 Large. Provider fal.ai, requested 1024x1024, n=1, seed=12166653711354006000, estimated cost $0.068157, model ID fal-ai/stable-diffusion-v35-large, generated Aug 14, 2026, 2:37 PM. Source

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

Prompt SHA 256: ed65cb4e6a51e2336f8b149ad4cbb16e020f07d527ad94362e719d2dcf91d94f

GPT Image 1.5, image SHA 256: 1c9ff59c32621c8b48b002dacb0ae66487262280c7c0a5b3d6456c81b248ebe9

GPT Image 1.5, request SHA 256: 7f1efc9531f1e6bed7d61a11c637add9fd892001aea2bb2781f8d89c79a2eb03

Stable Diffusion 3.5 Large, image SHA 256: 4eb0fd741ef42e6d3fee4b1280cdfb098a0a04009b139051963895892e349f1e

Stable Diffusion 3.5 Large, request SHA 256: 8b8014fc63f520d73ff57d3d8df5eae880097383167d44918b963f2830d09940

Product photography

Materials, reflections, and fine details

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 1.5. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-1.5, generated Aug 14, 2026, 12:16 PM. Source
Stable Diffusion 3.5 Large. Provider fal.ai, requested 1024x1024, n=1, seed=13346308312015933000, estimated cost $0.068157, model ID fal-ai/stable-diffusion-v35-large, generated Aug 14, 2026, 2:37 PM. Source

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

Prompt SHA 256: c67fd71e91304458a2feef5c91efa8677260fab0c0099a29a2c52cb579f5b2f3

GPT Image 1.5, image SHA 256: 67cfa2f324447027ceb5f82e9446d40a1395b46ce425890a74606b4d97731cf6

GPT Image 1.5, request SHA 256: c850843a298631a4076c5d2686de8e9d7223534f2b8ca53505f437862718bef9

Stable Diffusion 3.5 Large, image SHA 256: ea9a7401dee8859ec2cb96af9f7ca29a177ef2202f52f629d8ce9f5839bd4061

Stable Diffusion 3.5 Large, request SHA 256: 21e2c14d1b6315c30134e1390f531349608855d4ef1a87527e6d0ba57b2974ca

Typography and layout

Legible text, hierarchy, and composition

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 1.5. Provider Together, 1024x1024, n=1, seed=provider random, provider default steps, model ID openai/gpt-image-1.5, generated Aug 14, 2026, 1:59 PM. Source
Stable Diffusion 3.5 Large. Provider fal.ai, requested 1024x1024, n=1, seed=16632333557910893000, estimated cost $0.068157, model ID fal-ai/stable-diffusion-v35-large, generated Aug 14, 2026, 2:37 PM. Source

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

Prompt SHA 256: 2e012c3d14ee5dd02203a94f394ca80a5038cdebd6032b25a8b8a221f2255419

GPT Image 1.5, image SHA 256: 6a4cec86850fd371b65274198e2bf630b3cf36a096f9fb953d4bfe17c3222ebe

GPT Image 1.5, request SHA 256: 9c3a396aecfcaae3f8a25edbe5c8c44d0d33199d7ba29efda864d2b0221c3250

Stable Diffusion 3.5 Large, image SHA 256: e2994b1883a245b1c8c447957d121ea89d93aafff149029aa4e8507dbb769f9a

Stable Diffusion 3.5 Large, request SHA 256: 530872e813e988ba95025956a3743c6111120af963d90b8970a8bf3eced5547b

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

Compare GPT Image 1.5 and Stable Diffusion 3.5 Large with other important image models using the same benchmark and sample methodology.

View the complete image model comparison

GPT Image 1.5 vs. Stable Diffusion 3.5 Large

Change image model selection

General specifications and task-specific pricing

Shared capability benchmarks

Shared arena benchmarks

Sample images

Character and detail

Infographic

Product photography

Typography and layout

An Elo score is not a verdict on creativity

Related image model comparisons