Skip to main content

GPT Image 2vs.GPT Image 1.5

Is the extra cost worth it?

Probably not. In our own test both are on par. In the arena, GPT Image 2 leads. GPT Image 2 costs 1.6 times as much per image.

Test images

Same task, the first image of each. The task with the biggest difference is preselected.

GPT Image 2: Product photography, first image from the fixed test run
GPT Image 289 out of 100
GPT Image 1.5: Product photography, first image from the fixed test run
GPT Image 1.552 out of 100

Result for “Product photography”:GPT Image 2 ahead

Prompt given to both modelsCreate 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.

What the rating looks at: Materials, reflections, and fine details.

“Ahead”: better in all three rating rounds. “Slightly ahead”: a higher value, but not in every round.

Test protocol of the images

Each test attempt uses the same archived prompt and one request. The first output of that attempt counts. Repeats after technical test failures are disclosed for the affected images. Files are normalized to 1,024 × 1,024 pixels without cropping. Provider resolution, randomness, quality tier and prompt processing can still differ. The images are a practical visual test, not a reconstruction of the arena configuration.

Character and detail

GPT Image 2: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:13 PM. Source

GPT Image 1.5: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:17 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: 985a75b38642a9f0e5e0c0e0c67f1b7f67780d42bd2168fa24e9a7a7a94b0da6

GPT Image 2 image SHA 256: 82f6e2369ab4fb6efb0ef48feeff0d14f1d6087c64c35113a983456f0f0879ac

GPT Image 1.5 image SHA 256: 0dff52ccd39400ac66ac8870c9ea937512ff1474688e70118688f420c38c2296

Infographic

GPT Image 2: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:14 PM. Source

GPT Image 1.5: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:17 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: ed65cb4e6a51e2336f8b149ad4cbb16e020f07d527ad94362e719d2dcf91d94f

GPT Image 2 image SHA 256: 6c15466e258e2b14a8e862959cb6a67b4f7a23896ea5d42c5a5618d848847711

GPT Image 1.5 image SHA 256: 1c9ff59c32621c8b48b002dacb0ae66487262280c7c0a5b3d6456c81b248ebe9

Product photography

GPT Image 2: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:13 PM. Source

GPT Image 1.5: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 12:16 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: c67fd71e91304458a2feef5c91efa8677260fab0c0099a29a2c52cb579f5b2f3

GPT Image 2 image SHA 256: 471c71b30618afb1a8a67ab8363308d67f99c65ef6740ca72417806d06d3c346

GPT Image 1.5 image SHA 256: 67cfa2f324447027ceb5f82e9446d40a1395b46ce425890a74606b4d97731cf6

Typography and layout

GPT Image 2: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 1:59 PM. Source

GPT Image 1.5: together, 1024x1024, n=1, seed=provider random, provider default steps, generated Aug 14, 2026, 1:59 PM. Source

Protocol gradually-image-comparison-v1. Prompt SHA 256: 2e012c3d14ee5dd02203a94f394ca80a5038cdebd6032b25a8b8a221f2255419

GPT Image 2 image SHA 256: 434cf9939b9be2a2c2711e13d027c9438c47038bb32229ceeaf405d2e89d2afc

GPT Image 1.5 image SHA 256: 6a4cec86850fd371b65274198e2bf630b3cf36a096f9fb953d4bfe17c3222ebe

The numbers at a glance

Arena ratingfrom votes by users · higher is better
1,382

GPT Image 2

Rank 3 of 21

1,236

GPT Image 1.5

Rank 13 of 21

GraduallyBenchour own rating, 0 to 100 · higher is better
73

GPT Image 2

Rank 15 of 108

68

GPT Image 1.5

Rank 29 of 108

Price per imagedollars through the API · lower is better
$0.21

GPT Image 2

expensive

$0.13

GPT Image 1.5

expensive

Ranks among the 21 image models with an arena rating and the 108 with a GraduallyBench run; prices compared with 28 models with an API price.

GPT Image 2

Subscription
No OpenAI subscription includes this model. Through other providers from ≈ €6.36 a month (ElevenLabs Starter, with limited credits). 61 plans in Germany.
Via API
$0.21 per image, 100 images $21.10.

GPT Image 1.5

Subscription
No OpenAI subscription includes this model. Through other providers from ≈ €6.36 a month (ElevenLabs Starter, with limited credits). 42 plans in Germany.
Via API
$0.13 per image, 100 images $13.30.

Subscriptions only count plans that name exactly this model. Subscription prices include VAT, API prices are in dollars.

Arena rating (LMArena) from Nov 6, 2026 · GraduallyBench from Nov 3, 2026 · API prices from Nov 6, 2026 · Subscription database from Sep 29, 2026 (38 days old)

All data side by side

The better value is bold. For ratings only when the gap is clear.

KennzahlGPT Image 2GPT Image 1.5
Ratings
Arena ratingfrom votes by users, higher is better1,3821,236
Arena rankRank 3 of 21Rank 13 of 21
Arena votesmore votes, a safer rating106,737173,798
GraduallyBenchour own rating, 0 to 1007368
GraduallyBench by task
Typography and layoutLegible text, hierarchy, and composition7560
Product photographyMaterials, reflections, and fine details8952
Character and detailAnatomy, clothing, and spatial consistency6089
InfographicNumbers, labels, and visual organization6672
API price
Per imagelower is better$0.21$0.13
100 images$21.10$13.30
Profile
ReleasedApril 2026December 2025
Weightsopen means: runs locallyproprietaryproprietary
Lifecycleno noticeMarked as deprecated by OpenAI; removed from the API on December 1, 2026
AccessAPI, Web appAPI, Web app
Parametersnot publishednot published
Largest output4K1,536 × 1,024 pixels
Plans in Germanynaming exactly this model6142

API prices for one image at 1,024 × 1,024 pixels according to Artificial Analysis.

Sources: GPT Image 2 model source (verified Aug 15, 2026) · GPT Image 1.5 model source, status source (verified Aug 15, 2026)

Arena ratings by category

From blind votes on LMArena. “Ahead” means the lead is clear.

Arena rating per categoryfrom votes by users · higher is betterGPT Image 2GPT Image 1.5
Text to Image ArenaGPT Image 2: 1,188GPT Image 1.5: 1,124GPT Image 2 ahead
Image Editing ArenaGPT Image 2: 1,136GPT Image 1.5: 1,116level
Text to Image: 3D modelingGPT Image 2: 1,365GPT Image 1.5: 1,218GPT Image 2 ahead
Text to Image: Agentic image generationGPT Image 2: 1,338GPT Image 1.5: 1,237GPT Image 2 ahead
Text to Image: ArtGPT Image 2: 1,371GPT Image 1.5: 1,225GPT Image 2 ahead
Text to Image: CartoonGPT Image 2: 1,403GPT Image 1.5: 1,244GPT Image 2 ahead
Text to Image: Commercial designGPT Image 2: 1,386GPT Image 1.5: 1,238GPT Image 2 ahead
Text to Image: OverallGPT Image 2: 1,382GPT Image 1.5: 1,236GPT Image 2 ahead
Text to Image: PhotorealismGPT Image 2: 1,379GPT Image 1.5: 1,247GPT Image 2 ahead
Text to Image: PortraitsGPT Image 2: 1,426GPT Image 1.5: 1,259GPT Image 2 ahead
Text to Image: Text renderingGPT Image 2: 1,424GPT Image 1.5: 1,251GPT Image 2 ahead
Image Editing: Agentic image generationGPT Image 2: 1,460GPT Image 1.5: 1,360GPT Image 2 ahead
Image Editing: Multi-image editingGPT Image 2: 1,454GPT Image 1.5: 1,342GPT Image 2 ahead
Image Editing: Agentic multi-image editingGPT Image 2: 1,488GPT Image 1.5: 1,359GPT Image 2 ahead
Image Editing: OverallGPT Image 2: 1,460GPT Image 1.5: 1,368GPT Image 2 ahead

Artificial Analysis from Nov 6, 2026 · LMArena from Nov 6, 2026

All individual values with sources
RatingGPT Image 2GPT Image 1.5Source
Text to Image Arena1,1881,124Artificial AnalysisRetrieved Nov 6, 2026
Image Editing Arena1,1361,116Artificial AnalysisRetrieved Nov 6, 2026
Text to Image: 3D modeling1,3651,218LMArenaRetrieved Nov 6, 2026
Text to Image: Agentic image generation1,3381,237LMArenaRetrieved Nov 6, 2026
Text to Image: Art1,3711,225LMArenaRetrieved Nov 6, 2026
Text to Image: Cartoon1,4031,244LMArenaRetrieved Nov 6, 2026
Text to Image: Commercial design1,3861,238LMArenaRetrieved Nov 6, 2026
Text to Image: Overall1,3821,236LMArenaRetrieved Nov 6, 2026
Text to Image: Photorealism1,3791,247LMArenaRetrieved Nov 6, 2026
Text to Image: Portraits1,4261,259LMArenaRetrieved Nov 6, 2026
Text to Image: Text rendering1,4241,251LMArenaRetrieved Nov 6, 2026
Image Editing: Agentic image generation1,4601,360LMArenaRetrieved Nov 6, 2026
Image Editing: Multi-image editing1,4541,342LMArenaRetrieved Nov 6, 2026
Image Editing: Agentic multi-image editing1,4881,359LMArenaRetrieved Nov 6, 2026
Image Editing: Overall1,4601,368LMArenaRetrieved Nov 6, 2026

Published specialist tests

This pair shares 32 results from published research benchmarks.

Specialist testspublished research benchmarks · higher is betterGPT Image 2GPT Image 1.5
GenExam Mathematics, strictGPT Image 2: 50%GPT Image 1.5: 27%GPT Image 2 ahead
GenExam Physics, strictGPT Image 2: 80%GPT Image 1.5: 46%GPT Image 2 ahead
GenExam Chemistry, strictGPT Image 2: 70%GPT Image 1.5: 39%GPT Image 2 ahead
GenExam Biology, strictGPT Image 2: 89%GPT Image 1.5: 56%GPT Image 2 ahead
GenExam Geography, strictGPT Image 2: 85%GPT Image 1.5: 61%GPT Image 2 ahead
GenExam Computer science, strictGPT Image 2: 74%GPT Image 1.5: 36%GPT Image 2 ahead
GenExam Engineering, strictGPT Image 2: 79%GPT Image 1.5: 44%GPT Image 2 ahead
GenExam Economics, strictGPT Image 2: 83%GPT Image 1.5: 43%GPT Image 2 ahead
GenExam Music, strictGPT Image 2: 65%GPT Image 1.5: 29%GPT Image 2 ahead
GenExam History, strictGPT Image 2: 83%GPT Image 1.5: 51%GPT Image 2 ahead
GenExam overall, strictGPT Image 2: 75%GPT Image 1.5: 43%GPT Image 2 ahead
GenExam Mathematics, relaxedGPT Image 2: 85%GPT Image 1.5: 66%GPT Image 2 ahead
GenExam Physics, relaxedGPT Image 2: 96%GPT Image 1.5: 85%GPT Image 2 ahead
GenExam Chemistry, relaxedGPT Image 2: 92%GPT Image 1.5: 78%GPT Image 2 ahead
GenExam Biology, relaxedGPT Image 2: 98%GPT Image 1.5: 92%GPT Image 2 ahead
GenExam Geography, relaxedGPT Image 2: 98%GPT Image 1.5: 93%GPT Image 2 ahead
GenExam Computer science, relaxedGPT Image 2: 93%GPT Image 1.5: 76%GPT Image 2 ahead
GenExam Engineering, relaxedGPT Image 2: 97%GPT Image 1.5: 86%GPT Image 2 ahead
GenExam Economics, relaxedGPT Image 2: 98%GPT Image 1.5: 86%GPT Image 2 ahead
GenExam Music, relaxedGPT Image 2: 89%GPT Image 1.5: 71%GPT Image 2 ahead
GenExam History, relaxedGPT Image 2: 97%GPT Image 1.5: 91%GPT Image 2 ahead
GenExam overall, relaxedGPT Image 2: 94%GPT Image 1.5: 82%GPT Image 2 ahead
Qwen Image Bench qualityGPT Image 2: 59 / 100GPT Image 1.5: 55 / 100level
Qwen Image Bench aestheticsGPT Image 2: 68 / 100GPT Image 1.5: 61 / 100GPT Image 2 ahead
Qwen Image Bench alignmentGPT Image 2: 66 / 100GPT Image 1.5: 62 / 100level
Qwen Image Bench real-world fidelityGPT Image 2: 57 / 100GPT Image 1.5: 54 / 100level
Qwen Image Bench creative generationGPT Image 2: 75 / 100GPT Image 1.5: 66 / 100GPT Image 2 ahead
Qwen Image Bench overallGPT Image 2: 65 / 100GPT Image 1.5: 60 / 100GPT Image 2 ahead
GRADE reasoningGPT Image 2: 82 / 100GPT Image 1.5: 55 / 100GPT Image 2 ahead
GRADE consistencyGPT Image 2: 94 / 100GPT Image 1.5: 82 / 100GPT Image 2 ahead
GRADE readabilityGPT Image 2: 99 / 100GPT Image 1.5: 91 / 100GPT Image 2 ahead
GRADE accuracyGPT Image 2: 56 / 100GPT Image 1.5: 16 / 100GPT Image 2 ahead

GenExam from Aug 26, 2026 · Qwen Image Bench from Aug 26, 2026 · GRADE from Aug 26, 2026

All individual values with sources
RatingGPT Image 2GPT Image 1.5Source
GenExam Mathematics, strict50%27%GenExamRetrieved Aug 26, 2026
GenExam Physics, strict80%46%GenExamRetrieved Aug 26, 2026
GenExam Chemistry, strict70%39%GenExamRetrieved Aug 26, 2026
GenExam Biology, strict89%56%GenExamRetrieved Aug 26, 2026
GenExam Geography, strict85%61%GenExamRetrieved Aug 26, 2026
GenExam Computer science, strict74%36%GenExamRetrieved Aug 26, 2026
GenExam Engineering, strict79%44%GenExamRetrieved Aug 26, 2026
GenExam Economics, strict83%43%GenExamRetrieved Aug 26, 2026
GenExam Music, strict65%29%GenExamRetrieved Aug 26, 2026
GenExam History, strict83%51%GenExamRetrieved Aug 26, 2026
GenExam overall, strict75%43%GenExamRetrieved Aug 26, 2026
GenExam Mathematics, relaxed85%66%GenExamRetrieved Aug 26, 2026
GenExam Physics, relaxed96%85%GenExamRetrieved Aug 26, 2026
GenExam Chemistry, relaxed92%78%GenExamRetrieved Aug 26, 2026
GenExam Biology, relaxed98%92%GenExamRetrieved Aug 26, 2026
GenExam Geography, relaxed98%93%GenExamRetrieved Aug 26, 2026
GenExam Computer science, relaxed93%76%GenExamRetrieved Aug 26, 2026
GenExam Engineering, relaxed97%86%GenExamRetrieved Aug 26, 2026
GenExam Economics, relaxed98%86%GenExamRetrieved Aug 26, 2026
GenExam Music, relaxed89%71%GenExamRetrieved Aug 26, 2026
GenExam History, relaxed97%91%GenExamRetrieved Aug 26, 2026
GenExam overall, relaxed94%82%GenExamRetrieved Aug 26, 2026
Qwen Image Bench quality59 / 10055 / 100Qwen Image BenchRetrieved Aug 26, 2026
Qwen Image Bench aesthetics68 / 10061 / 100Qwen Image BenchRetrieved Aug 26, 2026
Qwen Image Bench alignment66 / 10062 / 100Qwen Image BenchRetrieved Aug 26, 2026
Qwen Image Bench real-world fidelity57 / 10054 / 100Qwen Image BenchRetrieved Aug 26, 2026
Qwen Image Bench creative generation75 / 10066 / 100Qwen Image BenchRetrieved Aug 26, 2026
Qwen Image Bench overall65 / 10060 / 100Qwen Image BenchRetrieved Aug 26, 2026
GRADE reasoning82 / 10055 / 100GRADERetrieved Aug 26, 2026
GRADE consistency94 / 10082 / 100GRADERetrieved Aug 26, 2026
GRADE readability99 / 10091 / 100GRADERetrieved Aug 26, 2026
GRADE accuracy56 / 10016 / 100GRADERetrieved Aug 26, 2026

More comparisons

Compare GPT Image 2 and GPT Image 1.5 with the models people compare them with most often.

Compare another matchup