Skip to content
IFAI

Model rankings

1090 models · 5 boards · 2 sources

Artificial AnalysisText-to-image arenaArenaText-to-image arena

Text-to-image arena · Anonymous head-to-head votes · Published March 26, 2026

Best open model
hunyuan-image-3.0Tencent1,151Score
Best from China
hunyuan-image-3.0Tencent1,151Score
Top three open models
1hunyuan-image-3.0Tencent · #13 overall1,151
2flux-2-devBlack Forest Labs · #14 overall1,148
3qwen-image-2512Alibaba · #18 overall1,136
This week's changes
qwen3.6-plus-previewCoding · #7New
gpt-5.4-mini-highCoding · #28New
Nemotron Cascade 2 30B A3BIntelligence · #78New
GPT-4Intelligence · #204New
Solar MiniIntelligence · #217New
RankChangeModelScore95% intervalVotes
1—gemini-3.1-flash-image-preview (nano-banana-2) [web-search]ClosedGoogle1,265±626,666
2—gpt-image-1.5-high-fidelityClosedOpenAI1,244±466,785
3—gemini-3-pro-image-preview-2k (nano-banana-pro)ClosedGoogle1,233±561,854
4—gemini-3-pro-image-preview (nano-banana-pro)ClosedGoogle1,232±582,543
5—mai-image-2ClosedMicrosoft AI1,190±611,001
6—reve-v1.5ClosedReve1,177±67,797
7—grok-imagine-imageClosedxAI1,173±453,997
8—flux-2-maxClosedBlack Forest Labs1,166±470,018
9—grok-imagine-image-proClosedxAI1,162±451,368
10—flux-2-flexClosedBlack Forest Labs1,158±4106,432
11—flux-2-proClosedBlack Forest Labs1,158±4101,679
12—gemini-2.5-flash-image-preview (nano-banana)ClosedGoogle1,155±3701,554
13—imagen-ultra-4.0-generate-001ClosedGoogle1,148±4387,965
14—seedream-4.5ClosedByteDance1,144±3109,184
15—seedream-4-2kClosedByteDance1,141±612,608
16—wan2.6-t2iClosedAlibaba1,135±451,327
17—imagen-4.0-generate-001ClosedGoogle1,132±3464,649
18—wan2.5-t2i-previewClosedAlibaba1,117±3144,593
19—seedream-4-falClosedByteDance1,117±611,847
20—gpt-image-1ClosedOpenAI1,115±3264,676
21—seedream-5.0-liteClosedByteDance1,115±525,118
22—seedream-4-high-res-falClosedByteDance1,113±4120,838
23—gpt-image-1-miniClosedOpenAI1,104±4110,053
24—recraft-v4ClosedRecraft1,099±618,350
25—mai-image-1ClosedMicrosoft AI1,093±494,380
26—seedream-3ClosedByteDance1,083±536,847
27—flux-1-kontext-maxClosedBlack Forest Labs1,075±365,635
28—flux-1-kontext-proClosedBlack Forest Labs1,059±3330,692
29—imagen-3.0-generate-002ClosedGoogle1,058±3360,180
30—ideogram-v3-qualityClosedIdeogram1,049±4115,098
31—photonClosedLuma AI1,035±4127,367
32—p-imageClosedPruna1,034±445,001
33—recraft-v3ClosedRecraft1,021±4191,824
34—runway-gen4ClosedRunway1,016±87,643
35—flux-1.1-proClosedBlack Forest Labs1,016±370,460
36—lucid-originClosedLeonardo.Ai1,013±3285,469
37—ideogram-v2ClosedIdeogram1,013±372,090
38—gemini-2.0-flash-preview-image-generationClosedGoogle975±3257,063
39—dall-e-3ClosedOpenAI968±4239,287

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 “—”.