Skip to content
IFAI

Model rankings

1412 models · 5 boards · 2 sources

‹
Week of August 10, 2026
September 21, 2026115 changesSeptember 14, 2026174 changesSeptember 7, 2026144 changesAugust 31, 202620 changesAugust 24, 202615 changesAugust 17, 202633 changesAugust 10, 2026116 changesAugust 3, 202613 changesJuly 27, 202639 changesJuly 20, 202618 changesJuly 13, 202611 changesJuly 6, 202626 changes
›
Artificial AnalysisText-to-image arenaArenaText-to-image arena

Text-to-image arena · Anonymous head-to-head votes · Published August 7, 2026

Best images
gpt-image-2 (medium)OpenAI1,380Score
Best open model
ideogram-4.0-qualityIdeogram1,206Score
Best from China
qwen-image-3.0-proAlibaba1,258Score
Most voted
gemini-2.5-flash-image-preview (nano-banana)Google822,940Votes
Top three by Score
1gpt-image-2 (medium)OpenAI1,380
2reve-2.1Reve1,302
3muse-imageMeta1,283
Top three open models
1ideogram-4.0-qualityIdeogram · #14 overall1,206
2Cosmos3-Super-Text2Image (Agentic)NVIDIA · #18 overall1,181
3hunyuan-image-3.0Tencent · #27 overall1,151
This week's changes
Qwen-Audio-3.0-TTS-PlusSpeech · #2New
Qwen3.8 MaxIntelligence · #5New
grok-4.6-highCoding · #5New
Muse Spark 1.2 (xhigh)Intelligence · #7New
gemini-3.7-flash-highCoding · #8New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
SortScoreVotes
RankChangeModelScore95% intervalVotes
1—gpt-image-2 (medium)ClosedOpenAI1,380±569,194
2—reve-2.1ClosedReve1,302±87,598
3—muse-imageClosedMeta1,283±714,510
4—reve-2.0ClosedReve1,270±614,650
5—gemini-3.1-flash-image (nano-banana-2) [web-search]ClosedGoogle1,263±527,910
6—qwen-image-3.0-proClosedAlibaba1,258±103,735
7—seedream-5.0-proClosedByteDance1,257±526,510
8—mai-image-2.5ClosedMicrosoft AI1,256±544,940
9—gemini-3.1-flash-lite-image (nano-banana-2-lite)ClosedGoogle1,251±616,419
10—gemini-3-pro-image-2k (nano-banana-pro)ClosedGoogle1,246±3138,756
11—gpt-image-1.5-high-fidelityClosedOpenAI1,239±3142,427
12—gemini-3-pro-image-preview (nano-banana-pro)ClosedGoogle1,232±582,835
13—grok-imagine-image-qualityClosedxAI1,228±448,786
14—ideogram-4.0-qualityOpenIdeogram1,206±527,891
15—qwen-image-2.0-pro-2026-06-22ClosedAlibaba1,191±612,026
16—uni-1.1-maxClosedLuma AI1,188±613,491
17—mai-image-2ClosedMicrosoft AI1,182±549,207
18—Cosmos3-Super-Text2Image (Agentic)OpenNVIDIA1,181±102,957
19—uni-1.1ClosedLuma AI1,180±526,927
20—grok-imagine-imageClosedxAI1,172±3217,461
21—recraft-v4.1-utility-proClosedRecraft1,169±112,519
22—flux-2-maxClosedBlack Forest Labs1,162±4117,464
23—grok-imagine-image-proClosedxAI1,161±493,711
24—Cosmos3-Super-Text2ImageOpenNVIDIA1,158±94,669
25—flux-2-flexClosedBlack Forest Labs1,156±3149,402
26—flux-2-proClosedBlack Forest Labs1,155±3175,196
27—reve-v1.5ClosedReve1,154±436,799
28—hunyuan-image-3.0OpenTencent1,151±3173,366
29—gemini-2.5-flash-image-preview (nano-banana)ClosedGoogle1,150±3822,940
30—imagen-ultra-4.0-generate-001ClosedGoogle1,148±4389,010
31—seedream-4.5ClosedByteDance1,146±3234,737
32—flux-2-devOpenBlack Forest Labs1,146±466,740
33—seedream-4-2kClosedByteDance1,140±712,639
34—seedream-5.0-liteClosedByteDance1,136±496,040
35—wan2.6-t2iClosedAlibaba1,136±3182,381
36—recraft-v4.1-proClosedRecraft1,130±112,682
37—imagen-4.0-generate-001ClosedGoogle1,129±3532,862
38—qwen-image-2512OpenAlibaba1,126±491,329
39—krea-2-mediumClosedKrea1,122±526,584
40—hidream-o1-imageOpenHiDream1,118±434,179
41—wan2.5-t2i-previewClosedAlibaba1,117±3230,234
42—seedream-4-falClosedByteDance1,116±711,911
43—gpt-image-1ClosedOpenAI1,115±3265,373
44—recraft-v4ClosedRecraft1,113±499,721
45—seedream-4-high-res-falClosedByteDance1,113±3175,579
46—krea-2-turboOpenKrea1,110±618,650
47—gpt-image-1-miniClosedOpenAI1,109±3164,065
48—krea-2-largeClosedKrea1,105±526,191
49—wan2.7-image-proClosedAlibaba1,103±528,745
50—wan2.7-imageClosedAlibaba1,100±529,087

Intelligence over time

Each maker's best model at the time on the Artificial Analysis Intelligence Index, by the month it was released.

Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi

Arena data comes from the leaderboard dataset it publishes on Hugging Face (CC BY 4.0). Only each board's overall leaderboard is used; scores and ranks are unchanged. dataset · CC BY 4.0

Artificial Analysis data comes from its API (artificialanalysis.ai). Scores are unchanged; ranks follow the scores. artificialanalysis.ai

Open or closed follows the license Arena gives; prices and Artificial Analysis intelligence show only where model names match, otherwise “—”.