Text-to-Image Arena Leaderboard

Name: Text-to-Image Arena Leaderboard
Creator: DataLearner
License: https://creativecommons.org/licenses/by/4.0/

The latest AI text-to-image model leaderboard based on Text-to-Image Arena anonymous user voting. Covers Elo scores, confidence intervals, and vote counts for GPT-Image, FLUX, Midjourney, DALL-E, and more.

Top Model

GPT-image-2 (medium)

Top Score

1,380

Model Count

Data version

2026年08月07日

Data source: LM Arena

About This Leaderboard

This leaderboard ranks AI text-to-image models by generation quality. Data comes from LMArena's Text-to-Image Arena track, evaluated through anonymous blind testing by real users.

Methodology Overview

Blind testing: Users submit text prompts, two anonymous models generate images, and users vote for the better result.

Elo scoring: Based on the Bradley-Terry model, scientifically measuring each model's relative strength in text-to-image generation.

Diverse generation scenarios: Covers photorealistic scenes, artistic illustration, product design, character creation, and more.

DataLearner provides in-depth analysis on top of the raw data, linking leaderboard models to the DataLearner model database so you can quickly access model details, API pricing, benchmark scores, and more.

Origin:All China

Leaderboard snapshot month:

Ranking Table

Rank	Model	Score	95% CI	Votes	Organization	License
	GPT-image-2 (medium)OpenAI	1,380	+/-5	69,194	OpenAI	Proprietary
	reve-2.1Reve	1,302	+/-8	7,598	Reve	Proprietary
	muse-imageMeta	1,283	+/-7	14,510	Meta	Proprietary
4	reve-2.0Reve	1,270	+/-6	14,650	Reve	Proprietary
5	gemini-3.1-flash-image (nano-banana-2) [web-search]Google	1,263	+/-5	27,910	Google	Proprietary
6	Qwen-Image-3.0阿里巴巴	1,258	+/-10	3,735	阿里巴巴	Proprietary
7	seedream-5.0-proBytedance	1,257	+/-5	26,510	Bytedance	Proprietary
8	mai-image-2.5 MicrosoftAI	1,256	+/-5	44,940	AI	Proprietary
9	gemini-3.1-flash-lite-image (nano-banana-2-lite)Google	1,251	+/-6	16,419	Google	Proprietary
10	gemini-3-pro-image-2k (nano-banana-pro)Google	1,246	+/-3	138,756	Google	Proprietary
11	GPT Image 1OpenAI	1,239	+/-3	142,427	OpenAI	Proprietary
12	Nano Banana ProGoogle Deep Mind	1,232	+/-5	82,835	Google Deep Mind	Proprietary
13	Grok Image 1.0xAI	1,228	+/-4	48,786	xAI	Proprietary
14	ideogram-4.0-qualityIdeogram	1,206	+/-5	27,891	Ideogram	Ideogram Open Model
15	qwen-image-2.0-pro-2026-06-22Alibaba	1,191	+/-6	12,026	Alibaba	Proprietary
16	uni-1.1-max LumaAI	1,188	+/-6	13,491	AI	Proprietary
17	MAI Image 2Microsoft Azure	1,182	+/-5	49,207	Microsoft Azure	Proprietary
18	Cosmos3-Super-Text2Image (Agentic)Nvidia	1,181	+/-10	2,957	Nvidia	OpenMDW-1.1
19	uni-1.1 LumaAI	1,180	+/-5	26,927	AI	Proprietary
20	Grok Image 1.0xAI	1,172	+/-3	217,461	xAI	Proprietary
21	recraft-v4.1-utility-proRecraft	1,169	+/-11	2,519	Recraft	Proprietary
22	flux-2-max Black ForestLabs	1,162	+/-4	117,464	Labs	Proprietary
23	Grok Image 1.0xAI	1,161	+/-4	93,711	xAI	Proprietary
24	Cosmos3-Super-Text2ImageNvidia	1,158	+/-9	4,669	Nvidia	OpenMDW-1.1
25	flux-2-flex Black ForestLabs	1,156	+/-3	149,402	Labs	Proprietary
26	flux-2-pro Black ForestLabs	1,155	+/-3	175,196	Labs	Proprietary
27	reve-v1.5Reve	1,154	+/-4	36,799	Reve	Proprietary
28	HunyuanImage-3.0-Instruct腾讯AI实验室	1,151	+/-3	173,366	腾讯AI实验室	tencent-hunyuan-community
29	Gemini 2.5 Flash Image PreviewGoogle Deep Mind	1,150	+/-3	822,940	Google Deep Mind	Proprietary
30	Imagen Ultra 4.0Google Deep Mind	1,148	+/-4	389,010	Google Deep Mind	Proprietary
31	Seedance 2.0字节跳动Seed团队	1,146	+/-3	234,737	字节跳动Seed团队	Proprietary
32	flux-2-dev Black ForestLabs	1,146	+/-4	66,740	Labs	flux-non-commercial-license
33	Seedream 4 2K字节跳动Seed团队	1,140	+/-7	12,639	字节跳动Seed团队	Proprietary
34	Seedream 5.0 Lite字节跳动Seed团队	1,136	+/-4	96,040	字节跳动Seed团队	Proprietary
35	Wan2.1-T2V-14B阿里巴巴	1,136	+/-3	182,381	阿里巴巴	Proprietary
36	recraft-v4.1-proRecraft	1,130	+/-11	2,682	Recraft	Proprietary
37	Imagen 4.0Google Deep Mind	1,129	+/-3	532,862	Google Deep Mind	Proprietary
38	Qwen-Image-2512阿里巴巴	1,126	+/-4	91,329	阿里巴巴	Apache 2.0
39	krea-2-mediumKrea	1,122	+/-5	26,584	Krea	Proprietary
40	hidream-o1-imageHiDream	1,118	+/-4	34,179	HiDream	MIT
41	Wan2.1-T2V-14B阿里巴巴	1,117	+/-3	230,234	阿里巴巴	Proprietary
42	Seedream 4 (FAL)字节跳动Seed团队	1,116	+/-7	11,911	字节跳动Seed团队	Proprietary
43	GPT Image 1OpenAI	1,115	+/-3	265,373	OpenAI	Proprietary
44	recraft-v4Recraft	1,113	+/-4	99,721	Recraft	Proprietary
45	Seedream 4 High Res (FAL)字节跳动Seed团队	1,113	+/-3	175,579	字节跳动Seed团队	Proprietary
46	Kling 2.5 Turbo昆仑万维	1,110	+/-6	18,650	昆仑万维	krea-2-community-license
47	GPT-Image-1-miniOpenAI	1,109	+/-3	164,065	OpenAI	Proprietary
48	krea-2-largeKrea	1,105	+/-5	26,191	Krea	Proprietary
49	Wan2.7 Image Pro阿里巴巴	1,103	+/-5	28,745	阿里巴巴	Proprietary
50	Wan2.7 Image阿里巴巴	1,100	+/-5	29,087	阿里巴巴	Proprietary
51	MAI Image 1Microsoft Azure	1,093	+/-4	94,700	Microsoft Azure	Proprietary
52	Seedream 3字节跳动Seed团队	1,082	+/-5	37,009	字节跳动Seed团队	Proprietary
53	Z-Image-Turbo阿里巴巴	1,081	+/-6	20,856	阿里巴巴	Apache 2.0
54	flux-1-kontext-max Black ForestLabs	1,074	+/-3	65,741	Labs	Proprietary
55	flux-2-klein-9b Black ForestLabs	1,070	+/-3	142,762	Labs	flux-non-commercial-license
56	Qwen Image阿里巴巴	1,060	+/-3	706,588	阿里巴巴	Apache 2.0
57	flux-1-kontext-pro Black ForestLabs	1,059	+/-3	331,629	Labs	Proprietary
58	Imagen 3.0 (002)Google Deep Mind	1,058	+/-3	360,739	Google Deep Mind	Proprietary
59	Qwen Image阿里巴巴	1,057	+/-3	84,835	阿里巴巴	Apache 2.0
60	ideogram-v3-qualityIdeogram	1,049	+/-4	115,578	Ideogram	Proprietary
61	photon LumaAI	1,035	+/-4	127,700	AI	Proprietary
62	p-image Proprietary	1,034	+/-4	103,942	—	—
63	flux-2-klein-4b Black ForestLabs	1,030	+/-3	144,584	Labs	Apache 2.0
64	runway-gen4Runway	1,025	+/-4	51,775	Runway	Proprietary
65	recraft-v3Recraft	1,021	+/-4	191,960	Recraft	Proprietary
66	flux-1.1-pro Black ForestLabs	1,016	+/-3	70,460	Labs	Proprietary
67	lucid-origin LeonardoAI	1,013	+/-3	285,797	AI	Proprietary
68	ideogram-v2Ideogram	1,013	+/-4	72,090	Ideogram	Proprietary
69	GLM-Image智谱AI	1,010	+/-9	4,626	智谱AI	MIT
70	Gemini 2.5 Flash Image PreviewGoogle Deep Mind	975	+/-3	257,623	Google Deep Mind	Proprietary
71	FLUX.1-Kontext-devBlack Forest Labs	969	+/-4	49,221	Black Forest Labs	Open
72	DALL·E3OpenAI	968	+/-4	239,491	OpenAI	Proprietary
73	flux-1-kontext-dev Black ForestLabs	940	+/-4	215,833	Labs	flux-1-dev-non-commercial-license
74	stable-diffusion-v35-large Open	938	+/-5	23,396	—	—
75	BAGEL字节跳动Seed团队	898	+/-6	12,404	字节跳动Seed团队	Apache 2.0

Data is for reference only. Official sources are authoritative. Click model names to view DataLearner model profiles.

2026-08 Market Signals

Current Best (SOTA)

GPT-Image-1.5 High-Fidelity (OpenAI)

Gemini 3 Pro Image Preview 2K (Google)

Gemini 3 Pro Image Preview (Google)

Best China Model

HunyuanImage-3.0 (腾讯)

Seedream-4.5 (字节跳动)

Qwen-Image-2512 (阿里)

Best Open Model

•Qwen-Image-2512 (阿里)
•Z-Image-Turbo (阿里)
•GLM-Image (智谱)

FAQ

What is the difference between text-to-image and image editing?

Text-to-image creates a new image from a prompt. Image editing modifies an existing image, which is better for local changes, style transfer, and production refinements.

Which models are suitable for commercial poster design?

For commercial posters, prioritize models with strong text rendering, controllable composition, high-resolution output, and license terms that fit your use case. The top-ranked model may not be the best option if typography or brand control matters most.

What is prompt engineering?

Prompt engineering means structuring the text input to guide the generated image. Clear descriptions of subject, style, lighting, composition, and constraints usually improve quality and alignment.

How large is the gap between open and closed image models?

The gap has narrowed, especially for customizable open models. Closed models can still lead on instruction following, typography, and detail consistency, while open models are attractive for local deployment, tuning, and cost control.