What is the best AI text-to-image model?

Based on blind community voting, Nano Banana 2 is currently the #1 ranked AI text-to-image model with an Elo rating of 1279. Rankings update in real time as new votes come in.

How are AI text-to-image models ranked on Lumenfall?

Lumenfall Arena ranks AI models through blind community voting. In each matchup, two models generate from the same prompt and voters pick the better result without seeing model names. Votes are processed using TrueSkill, a Bayesian rating algorithm developed by Microsoft Research, that produces a single Elo score reflecting each model's relative quality.

What is an Elo rating for AI models?

An Elo rating is a numerical score representing a model's skill relative to other models. Under the hood, Lumenfall uses TrueSkill, which tracks two values per model: mu (estimated skill) and sigma (uncertainty). The displayed Elo is calculated as 1000 + 10 x (mu - 3*sigma), a conservative lower bound. A model must prove itself consistently across many matchups to earn a high rating.

Best AI Models for Aesthetics

Aesthetics · Text-to-Image

Elo rankings from blind votes across 11 challenges in this category.

ImagineArt 1.5 (Preview)

Grok Imagine Image Pro

Highlights

10 images

Price · Performance

Best models, by

Every ranked model plotted by price against its Elo. The upper-left frontier is the best value: top quality at the lowest cost.

Best Text-to-Image Models by Price

1 model without pricing omitted.

Best AI Models for Aesthetics

45 models ranked · Last update: July 3, 2026 10:20 AM

#	Model	Price	¢/img	Speed	Elo
1	Nano Banana 2 Google	$$	2.2¢		1279
2	GPT Image 1.5 OpenAI	$	0.9¢		1278
3	Nano Banana Pro Google	$$$	6.7¢		1277
4	GPT Image 2 OpenAI	$$	3.9¢		1275
5	FLUX.2 [dev] Flash fal	$	0.5¢		1261
6	FLUX.2 [dev] Turbo fal	$	0.8¢		1261
7	ImagineArt 1.5 (Preview) Vyro AI	$$	3¢	—	1259
8	Recraft V4 Recraft AI	$$	4¢		1257
9	FLUX.2 [max] Black Forest Labs	$$	3¢		1255
10	Seedream 4.5 ByteDance	$$	4¢		1255

11	FLUX.1 [dev] Black Forest Labs	$$	2.5¢		1249
12	FLUX.2 [pro] Black Forest Labs	$$	1.5¢		1249
13	GPT Image 1 Mini OpenAI	$	0.5¢		1244
14	Z-Image Turbo Alibaba	$	0.5¢		1242
15	FLUX.2 [dev] Black Forest Labs	$$	1.2¢		1241
16	FLUX.2 [flex] Black Forest Labs	$$$	6¢		1240
17	Seedream 4.0 ByteDance	$$	3¢		1239
18	Grok Imagine Image Pro xAI	$$$	7¢		1239
19	Recraft V4 Pro Recraft AI	$$$$	25¢		1238
20	Seedream 5.0 Lite ByteDance	$$	3.5¢		1232
21	Nano Banana Google	$$	3.9¢		1229
22	Lucid Origin Leonardo AI	$$	2¢	—	1228
23	Wan 2.6 Alibaba	$$	3¢		1224
24	Qwen Image Max Alibaba	$$$	7.5¢		1223
25	Grok Imagine Image xAI	$$	2¢		1220
26	FLUX.2 [klein] 9B Black Forest Labs	$	0.6¢		1219
27	Imagen 4.0 Ultra Generate 001 Google	$$$	6¢		1218
28	Qwen Image 2512 Alibaba	$$	2¢		1215
29	Stable Diffusion 3.5 Large Stability AI	$$$	6.5¢		1214
30	Wan 2.7 Pro Alibaba	$$$	7.5¢		1213

Google’s Nano Banana 2 leads the aesthetic rankings with 1279 Elo and a 76.3% win rate, narrowly outpacing OpenAI’s GPT Image 1.5 by a single Elo point. While GPT Image 2 maintains the highest specific win rate at 85.1%, the budget-friendly FLUX.2 [dev] Flash delivers a competitive 1261 Elo at approximately 13% of the price of premium competitors.

Rivalries

Aggregate head-to-head across the arena

FLUX.2 [pro] Grok Imagine Image

50% 6 challenges 50%

GPT Image 1.5 FLUX.2 [dev] Flash

44% 3 challenges 44%

Seedream 4.5 Nano Banana Pro

38% 4 challenges 50%

ImagineArt 1.5 (Preview) FLUX.2 [max]

38% 5 challenges 50%

Think a model is ranked wrong? Cast your vote

Highlighted challenges

GPT Image 2 is first to make it GPT Image 2

Text-to-Image Art Photorealism

The Reversed Rodeo

This competition tests how well AI image models truly understand language versus how much they rely on visual habits from their training data. The prompt is deliberately simple on the surface but devilishly hard in practice. Most models default to the familiar trope of an astronaut riding a horse. By forcing the reversal, we measure three critical capabilities that separate good models from great ones: Strict instruction following (including negations) Accurate subject-object relationships and spatial hierarchy Resistance to strong dataset biases

Text-to-Image 3D Imaging & Modeling Photorealism

Geometric Composition

A spatial-reasoning test. Each object has a precise relationship (inside, on top, behind, seen through the glass), so it measures whether a model follows explicit placement instructions and handles transparency and refraction rather than just approximating the scene.

View all challenges

Keep the arena honest

Cast your vote

Pick winners in blind matchups. Every vote nudges the Elo and shapes these rankings.

Cast Your Vote

Suggest a prompt

Got an idea worth testing? Submit a prompt and watch the models battle it out.

Vote Now

Models · slot A

Aesthetics · Text-to-Image

Highlights

Best models, by

Best Text-to-Image Models by Price

Best Text-to-Image Models by Speed

Best AI Models for Aesthetics

Rivalries

Highlighted challenges

The Reversed Rodeo

Geometric Composition

Cast your vote

Suggest a prompt