Skip to content
IFAI

Model rankings

1542 models · 5 boards · 2 sources

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

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

Best images
gpt-image-2.5-sunburstOpenAI1,425Score
Best open model
qwen-image-2.1Alibaba1,223Score
Best from China
seedream-5.0-proByteDance1,255Score
Top three by Score
2gpt-image-2.5-flareOpenAI1,398
Top three open models
1qwen-image-2.1Alibaba · #18 overall1,223
2ideogram-4.0-qualityIdeogram · #19 overall1,205
3Cosmos3-Super-Text2Image (Agentic)NVIDIA · #26 overall1,165
This week's changes
Claude Opus 5.5 (Max)Coding · #1New
Eleven v4 TurboSpeech · #1New
GPT-6 Astra (Max)Coding · #2New
GPT-6.1 Sol (Max)Coding · #4New
RankChangeModelScore95% intervalVotes
1—gpt-image-2.5-sunburstClosedOpenAI1,425±717,752
2—gpt-image-2.5-flareClosedOpenAI1,398±716,889
3—gpt-image-2 (medium)ClosedOpenAI1,383±493,334
4—grok-imagine-image-2.0 (canvas)ClosedxAI · 1 more variant1,334±123,113
5—mai-image-2.6ClosedMicrosoft AI1,332±525,091
6—gemini-nano-banana-2.1ClosedGoogle1,328±95,310
7—reve-2.1ClosedReve1,302±88,249
8—muse-imageClosedMeta1,274±542,262
9—reve-2.0ClosedReve1,269±615,773
10—gemini-3.1-flash-image (nano-banana-2) [web-search]ClosedGoogle1,262±460,515
11—mai-image-2.5ClosedMicrosoft AI1,254±461,265
12—gemini-3.1-flash-lite-image (nano-banana-2-lite)ClosedGoogle1,250±617,804
13—gemini-3-pro-image-2k (nano-banana-pro)ClosedGoogle1,248±3175,734
14—gpt-image-1.5-high-fidelityClosedOpenAI1,237±3169,927
15—gemini-3-pro-image-preview (nano-banana-pro)ClosedGoogle1,234±586,464
16—ideogram-4.0-qualityOpenIdeogram1,205±460,591
17—uni-1.1-maxClosedLuma AI1,189±614,072
18—uni-1.1ClosedLuma AI1,184±446,699
19—mai-image-2ClosedMicrosoft AI1,184±551,475
20—grok-imagine-imageClosedxAI1,170±3272,085
21—Cosmos3-Super-Text2Image (Agentic)OpenNVIDIA · 1 more variant1,165±710,805
22—grok-imagine-image-proClosedxAI1,162±499,455
23—reve-v1.5ClosedReve1,155±439,700
24—gemini-2.5-flash-image-preview (nano-banana)ClosedGoogle1,150±2890,539
25—imagen-ultra-4.0-generate-001ClosedGoogle1,148±4388,736
26—imagen-4.0-generate-001ClosedGoogle1,129±3539,732
27—krea-2-mediumClosedKrea1,124±458,098
28—gpt-image-1ClosedOpenAI1,116±3268,453
29—krea-2-turboOpenKrea1,113±451,552
30—krea-2-largeClosedKrea1,112±456,888
31—gpt-image-1-miniClosedOpenAI1,110±3171,058
32—mai-image-1ClosedMicrosoft AI1,093±498,410
33—imagen-3.0-generate-002ClosedGoogle1,058±3359,772
34—ideogram-v3-qualityClosedIdeogram1,048±4118,538
35—photonClosedLuma AI1,036±4130,160
36—runway-gen4ClosedRunway1,024±454,361
37—ideogram-v2ClosedIdeogram1,013±472,090
38—lucid-originClosedLeonardo.Ai1,013±3287,341
39—gemini-2.0-flash-preview-image-generationClosedGoogle975±3257,227
40—dall-e-3ClosedOpenAI968±4238,743

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 “—”.