Skip to content
IFAI

Model rankings

521 models · 3 boards · 2 sources

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

Text-to-image arena · Anonymous head-to-head votes · Published August 28, 2025

Best open model
qwen-image-prompt-extendAlibaba1,081Score
Best from China
qwen-image-prompt-extendAlibaba1,081Score
Most voted
recraft-v3Recraft290,850Votes
Top three by Score
3gpt-image-1OpenAI1,124
Top three open models
1qwen-image-prompt-extendAlibaba · #5 overall1,081
2qwen-imageAlibaba · #9 overall1,051
3flux-1-dev-fp8Black Forest Labs · #19 overall968
This week's changes
Exaone 4.0 1.2B (Reasoning)Intelligence · #81New
Qwen3 4B 2507 (Reasoning)Intelligence · #26Up 59
Qwen3 0.6B (Reasoning)Intelligence · #148Up 10
RankChangeModelScore95% intervalVotes
1—gemini-2.5-flash-image-preview (nano-banana)ClosedGoogle1,145±3256,820
2—imagen-ultra-4.0-generate-001ClosedGoogle1,133±2248,036
3—gpt-image-1ClosedOpenAI1,124±3150,073
4—imagen-4.0-generate-001ClosedGoogle1,117±2251,756
5—imagen-3.0-generate-002ClosedGoogle1,061±3290,278
6—photonClosedLuma AI1,034±3219,075
7—lucid-originClosedLeonardo.Ai1,033±2141,748
8—ideogram-v3-qualityClosedIdeogram1,031±3200,505
9—ideogram-v2ClosedIdeogram1,010±3277,638
10—gemini-2.0-flash-preview-image-generationClosedGoogle987±3177,095
11—dall-e-3ClosedOpenAI978±3253,607

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 “—”.