Valumigo

LMArena · Text-to-image

We regularly fetch and display public benchmark data: LMArena user voting rankings, Epoch AI test scores, and OpenRouter usage rankings. Each leaderboard shows both its publication date and the date we retrieved it. These scores are not assigned by this site.

LMArena scores come from people comparing two models' answers side by side and voting for the better one, using an Elo-based system. If score differences fall within the confidence intervals, treat the models as roughly comparable.

How to read this leaderboard

What it measures: An image generation ranking based on LMArena votes comparing two images generated from the same prompt with model names hidden.

How to read the score: A relative score reflecting which images users preferred. Overlapping confidence intervals alone do not establish that two models have equal capability.

Caveats: This is not a composite metric that separately evaluates size, speed and price. Check your required conditions, such as rendering Korean (Hangul) characters, and costs separately.

Insights from this ranking

  • The 95% confidence intervals for 1st-place gpt-image-2.5-sunburst and 2nd-place gpt-image-2.5-flare do not overlap. This aggregation provides relatively clear evidence that the 1st-place model leads.
  • OpenAI has the most models among the top 10, with 3.
  • The 1st-place model received 10,884 votes, and this leaderboard includes 50 models.

Text-to-image LMArena

Published 2026-09-24 · Retrieved 2026-10-04

112011901260133014001470
Ogpt-image-2.5-sunburst
1424
Ogpt-image-2.5-flare
1401
Ogpt-image-2 (medium)
1383
Mmai-image-2.6
1335
Rreve-2.1
1302
xgrok-imagine-image-2.0 (low)
1301
Metamuse-image
1276
Rreve-2.0
1269
Googlegemini-3.1-flash-image (nano-banana-2) [web-search]
1261
ByteDanceseedream-5.0-pro
1256
Alibabaqwen-image-3.0-pro
1256
Mmai-image-2.5
1254
Googlegemini-3.1-flash-lite-image (nano-banana-2-lite)
1250
Googlegemini-3-pro-image-2k (nano-banana-pro)
1247
Ogpt-image-1.5-high-fidelity
1238
Googlegemini-3-pro-image-preview (nano-banana-pro)
1234
Alibabaqwen-image-2.1
1228
Iideogram-4.0-quality
1205
Alibabaqwen-image-2.0-pro-2026-06-22
1192
Luni-1.1-max
1189

Dots show scores; horizontal lines show 95% confidence intervals. Overlapping intervals suggest similar performance. The chart shows only the highest-scoring reasoning setting, such as high or max, for each model. The table shows rankings for all settings.

Top ranking by company

  • OOpenAIgpt-image-2.5-sunburst#1
  • MMicrosoft AImai-image-2.6#4
  • RRevereve-2.1#5
  • xxAIgrok-imagine-image-2.0 (low)#6
  • MetaMetamuse-image#7
  • GoogleGooglegemini-3.1-flash-image (nano-banana-2) [web-search]#9
  • ByteDanceByteDanceseedream-5.0-pro#10
  • AlibabaAlibabaqwen-image-3.0-pro#11
  • IIdeogramideogram-4.0-quality#18
View table (top 50)
RankModelDeveloperScore95% confidence intervalVotes
1#1Ogpt-image-2.5-sunburstOpenAI14241416–143110,884
2#2Ogpt-image-2.5-flareOpenAI14011393–140910,058
3#3Ogpt-image-2 (medium)OpenAI13831379–138788,744
4#4Mmai-image-2.6Microsoft AI13351329–134118,625
5#5Rreve-2.1Reve13021295–13108,305
6#6xgrok-imagine-image-2.0 (low)xAI13011293–13087,257
7#7Metamuse-imageMeta12761271–128135,422
8#8Rreve-2.0Reve12691263–127515,802
9#9Googlegemini-3.1-flash-image (nano-banana-2) [web-search]Google12611257–126553,228
10#10ByteDanceseedream-5.0-proByteDance12561252–126085,690
11#11Alibabaqwen-image-3.0-proAlibaba12561249–126211,520
12#12Mmai-image-2.5Microsoft AI12541250–125861,448
13#13Googlegemini-3.1-flash-lite-image (nano-banana-2-lite)Google12501244–125617,851
14#14Googlegemini-3-pro-image-2k (nano-banana-pro)Google12471244–1250168,447
15#15Ogpt-image-1.5-high-fidelityOpenAI12381235–1242165,437
16#16Googlegemini-3-pro-image-preview (nano-banana-pro)Google12341229–123986,451
17#17Alibabaqwen-image-2.1Alibaba12281219–12374,445
18#18Iideogram-4.0-qualityIdeogram12051201–120953,371
19#19Alibabaqwen-image-2.0-pro-2026-06-22Alibaba11921186–119713,123
20#20Luni-1.1-maxLuma AI11891183–119514,083
21#21Mmai-image-2Microsoft AI11841179–118851,477
22#22Luni-1.1Luma AI11831179–118841,412
23#23xgrok-imagine-imagexAI11701168–1173262,132
24#24Rrecraft-v4.1-utility-proRecraft11691158–11802,618
25#25NVIDIACosmos3-Super-Text2Image (Agentic)NVIDIA11651158–117210,856
26#26xgrok-imagine-image-proxAI11621158–116699,444
27#27Bflux-2-maxBlack Forest Labs11621159–1166123,187
28#28Bflux-2-flexBlack Forest Labs11571153–1160156,912
29#29Rreve-v1.5Reve11551151–115939,755
30#30Bflux-2-proBlack Forest Labs11541151–1157205,540
31#31NVIDIACosmos3-Super-Text2ImageNVIDIA11521146–115818,581
32#32Googlegemini-2.5-flash-image-preview (nano-banana)Google11501148–1153878,451
33#33Thunyuan-image-3.0Tencent11501146–1153180,550
34#34Googleimagen-ultra-4.0-generate-001Google11481143–1152388,772
35#35ByteDanceseedream-4.5ByteDance11471145–1150287,128
36#36Bflux-2-devBlack Forest Labs11461141–115078,080
37#37ByteDanceseedream-4-2kByteDance11401133–114712,566
38#38ByteDanceseedream-5.0-liteByteDance11381135–1141123,178
39#39Alibabawan2.6-t2iAlibaba11371134–1139223,535
40#40Rrecraft-v4.1-proRecraft11331122–11432,795
41#41Googleimagen-4.0-generate-001Google11291126–1132539,788
42#42Alibabaqwen-image-2512Alibaba11251122–112998,363
43#43Kkrea-2-mediumKrea11231119–112750,800
44#44ByteDanceseedream-4-falByteDance11171110–112311,842
45#45Alibabawan2.5-t2i-previewAlibaba11161114–1119271,922
46#46Hhidream-o1-imageHidream11161112–112059,511
47#47Ogpt-image-1OpenAI11161113–1119268,466
48#48Rrecraft-v4Recraft11151112–1118129,674
49#49ByteDanceseedream-4-high-res-falByteDance11131110–1117182,669
50#50Kkrea-2-turboKrea11131108–111743,913

Data: LMArena leaderboard dataset (CC BY 4.0). Changes: Top 50 entries per leaderboard; scores are rounded. https://huggingface.co/datasets/lmarena-ai/leaderboard-datasetCompany logos are trademarks of their respective owners and are used only for identification (icons: Simple Icons).Data: Epoch AI — Capabilities & benchmarking (CC BY 4.0). https://epoch.ai/benchmarksSource: OpenRouter (openrouter.ai/rankings), as of 2026-10-04. Licensed under CC BY 4.0. https://openrouter.ai/rankings