Skip to content
IFAI

Model rankings

1125 models · 5 boards · 2 sources

‹
Week of April 6, 2026
May 18, 202616 changesMay 11, 202665 changesMay 4, 202616 changesApril 27, 202624 changesApril 20, 202627 changesApril 13, 202637 changesApril 6, 202641 changesMarch 30, 2026230 changesMarch 23, 202626 changesMarch 16, 202622 changesMarch 9, 202665 changesMarch 2, 202663 changes
›
Artificial AnalysisText-to-image arenaArenaText-to-image arena

Text-to-image arena · Anonymous head-to-head votes · Updated April 12, 2026

Best images
GPT Image 1.5 (high)OpenAI1,265Score
Best open model
FLUX.2 [dev] TurboFal1,163Score
Best from China
Seedream 4.0ByteDance Seed1,185Score
Top three by Score
1GPT Image 1.5 (high)OpenAI1,265
2Nano Banana 2 (Gemini 3.1 Flash Image Preview)Google1,258
3Riverflow 2.0Sourceful1,254
Top three open models
1FLUX.2 [dev] TurboFal · #15 overall1,163
2Qwen Image Max 2512Alibaba · #16 overall1,150
3FLUX.2 [dev]Black Forest Labs · #20 overall1,148
This week's changes
HappyHorse-1.0Video · #1New
glm-5.1Coding · #2New
Dreamina Seedance 2.0 720pVideo · #2New
Eleven v3Speech · #2New
Speech 2.8 HDSpeech · #4New
AllOpenClosed
All regionsUSChina
One row per modelAll variants
RankChangeModelScore95% interval
1—GPT Image 1.5 (high)ClosedOpenAI1,265±11
2—Nano Banana 2 (Gemini 3.1 Flash Image Preview)ClosedGoogle1,258±9
3—Riverflow 2.0ClosedSourceful1,254±11
4—Nano Banana Pro (Gemini 3 Pro Image)ClosedGoogle1,215±11
5—FLUX.2 [max]ClosedBlack Forest Labs1,200±11
6—Seedream 4.0ClosedByteDance Seed1,185±7
7—FLUX.2 [pro]ClosedBlack Forest Labs1,182±12
8—FLUX.2 [flex]ClosedBlack Forest Labs1,180±11
9—Imagen 4 Ultra Preview 0606ClosedGoogle1,174±11
10▲ 1grok-imagine-imageClosedxAI1,172±8
11▼ 1Seedream 4.5ClosedByteDance Seed1,171±9
12—Imagen 4 Preview 0606ClosedGoogle1,170±12
13—Imagen 4 UltraClosedGoogle1,170±11
14—Nano Banana (Gemini 2.5 Flash Image)ClosedGoogle1,163±7
15—Seedream 3.0ClosedByteDance Seed1,149±10
16—Wan 2.5 PreviewClosedAlibaba1,148±12
17—ImagineArt 1.5 PreviewClosedImagineArt1,148±8
18—grok-imagine-image-proClosedxAI1,138±8
19—Wan 2.6 ImageClosedAlibaba1,133±16
20—Wan2.6 Text to ImageClosedAlibaba1,133±9
21—Recraft V4 ProClosedRecraft1,132±9
22—GPT Image 1 (high)ClosedOpenAI1,128±11
23—Vivago 2.1ClosedHiDream1,126±11
24—Kolors 2.1ClosedKlingAI1,124±11
25—Recraft V4ClosedRecraft1,121±9
26—Seedream 5.0 LiteClosedByteDance1,119±10
27—FLUX.1 Kontext [max]ClosedBlack Forest Labs1,115±9
28—Vidu Q2ClosedVidu1,113±9
29—Lucid Origin UltraClosedLeonardo.Ai1,111±11
30—Eigen ImageClosedEigen AI1,107±8
31—Lucid Origin FastClosedLeonardo.Ai1,106±11
32—Vivago 2.0ClosedHiDream1,103±12
33—Imagen 3 (v002)ClosedGoogle1,103±10
34—Imagen 4 StandardClosedGoogle1,097±11
35—Reve Image (Halfmoon)ClosedReve1,090±13
36—FLUX1.1 [pro] UltraClosedBlack Forest Labs1,088±12
37—Imagen 4 Fast Preview 0606ClosedGoogle1,082±11
38—FLUX.1 Kontext [pro]ClosedBlack Forest Labs1,082±12
39—Dreamina 3.1ClosedByteDance1,080±11
40—Ideogram 3.0ClosedIdeogram1,076±13
41—FLUX1.1 [pro]ClosedBlack Forest Labs1,076±9
42—Recraft V3ClosedRecraft1,073±9
43—GPT Image 1 Mini (medium)ClosedOpenAI1,071±11
44—FLUX.1 [pro]ClosedBlack Forest Labs1,068±8
45▲ 2P-ImageClosedPruna AI1,067±13
46▼ 1Midjourney v7 AlphaClosedMidjourney1,066±13
47▼ 1Qwen Image Plus 2601ClosedAlibaba1,066±8
48—Imagen 4 FastClosedGoogle1,065±11
49—Luma PhotonClosedLuma Labs1,064±10
50—Ideogram v2ClosedIdeogram1,057±8

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 “—”.