加载中...
加载中...
基于 Text-to-Image Arena 用户匿名投票的最新AI文生图模型排行榜,涵盖各模型的 Elo 得分、95% 置信区间、投票量、机构与许可证。
Top Model
gpt-image-1.5-high-fidelity
Top Score
1,248
Model Count
46
Data version
2026年02月12日
Data source: LM Arena
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.
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.
Chart Source: DataLearnerAI · Data Source: LMArena
| Rank | Model | Score | 95% CI | Votes | Organization | License |
|---|---|---|---|---|---|---|
| 1 | gpt-image-1.5-high-fidelity | 1,248 | +5/-5 | 41,719 | OpenAI | Proprietary |
| 2 | gemini-3-pro-image-preview-2k (nano-banana-pro) | 1,237 | +5/-5 | 41,547 | Proprietary | |
| 3 | gemini-3-pro-image-preview (nano-banana-pro) | 1,233 | +5/-5 | 83,656 | Proprietary | |
| 4 | grok-imagine-image | 1,174 | +6/-6 | 11,585 | xAI | Proprietary |
| 5 | flux-2-max | 1,169 | +4/-4 | 47,290 | Black Forest Labs | Proprietary |
| 6 | grok-imagine-image-pro | 1,166 | +6/-6 | 13,012 | xAI | Proprietary |
| 7 | flux-2-flex | 1,158 | +4/-4 | 66,495 | Black Forest Labs | Proprietary |
| 8 | gemini-2.5-flash-image-preview (nano-banana) | 1,157 | +3/-3 | 653,540 | Proprietary | |
| 9 | flux-2-pro | 1,156 | +4/-4 | 78,124 | Black Forest Labs | Proprietary |
| 10 | hunyuan-image-3.0 | 1,151 | +3/-3 | 158,902 | Tencent | Proprietary |
| 11 | flux-2-dev | 1,150 | +5/-5 | 39,481 | Black Forest Labs | Proprietary |
| 12 | imagen-ultra-4.0-generate-001 | 1,149 | +4/-4 | 390,174 | Proprietary | |
| 13 | seedream-4-2k | 1,141 | +6/-6 | 12,571 | Bytedance | Proprietary |
| 14 | seedream-4.5 | 1,141 | +4/-4 | 49,398 | Bytedance | Proprietary |
| 15 | qwen-image-2512 | 1,139 | +5/-5 | 29,166 | Alibaba | Apache 2.0 |
| 16 | imagen-4.0-generate-001 | 1,135 | +3/-3 | 441,159 | Proprietary | |
| 17 | wan2.6-t2i | 1,126 | +6/-6 | 11,041 | Alibaba | Proprietary |
| 18 | seedream-4-fal | 1,119 | +6/-6 | 11,754 | Bytedance | Proprietary |
| 19 | wan2.5-t2i-preview | 1,117 | +4/-4 | 102,970 | Alibaba | Proprietary |
| 20 | gpt-image-1 | 1,115 | +3/-3 | 252,289 | OpenAI | Proprietary |
| 21 | seedream-4-high-res-fal | 1,114 | +4/-4 | 103,463 | Bytedance | Proprietary |
| 22 | gpt-image-1-mini | 1,100 | +4/-4 | 85,364 | OpenAI | Proprietary |
| 23 | mai-image-1 | 1,094 | +4/-4 | 74,023 | Microsoft AI | Proprietary |
| 24 | seedream-3 | 1,084 | +5/-5 | 36,622 | Bytedance | Proprietary |
| 25 | z-image-turbo | 1,083 | +7/-7 | 7,577 | Alibaba | Apache 2.0 |
| 26 | flux-1-kontext-max | 1,076 | +3/-3 | 66,184 | Black Forest Labs | Proprietary |
| 27 | flux-2-klein-9b | 1,065 | +4/-4 | 27,524 | Black Forest Labs | flux-non-commercial-license |
| 28 | qwen-image-prompt-extend | 1,061 | +3/-3 | 579,074 | Alibaba | Apache 2.0 |
| 29 | flux-1-kontext-pro | 1,060 | +3/-3 | 333,004 | Black Forest Labs | Proprietary |
| 30 | imagen-3.0-generate-002 | 1,059 | +3/-3 | 361,579 | Proprietary | |
| 31 | qwen-image | 1,058 | +2/-2 | 85,994 | Alibaba | Apache 2.0 |
| 32 | p-image | 1,053 | +5/-5 | 17,651 | Phuna | Proprietary |
| 33 | ideogram-v3-quality | 1,050 | +4/-4 | 115,778 | Ideogram | Proprietary |
| 34 | photon | 1,037 | +4/-4 | 127,980 | Luma AI | Proprietary |
| 35 | flux-2-klein-4b | 1,021 | +4/-4 | 27,669 | Black Forest Labs | Apache 2.0 |
| 36 | recraft-v3 | 1,021 | +3/-3 | 178,884 | Recraft | Proprietary |
| 37 | flux-1.1-pro | 1,017 | +3/-3 | 70,353 | Black Forest Labs | Proprietary |
| 38 | lucid-origin | 1,015 | +3/-3 | 287,569 | Leonardo AI | Proprietary |
| 39 | ideogram-v2 | 1,015 | +3/-3 | 72,113 | Ideogram | Proprietary |
| 40 | glm-image | 1,013 | +9/-9 | 4,680 | Z.ai | MIT |
| 41 | gemini-2.0-flash-preview-image-generation | 978 | +3/-3 | 258,646 | Proprietary | |
| 42 | flux-1-dev-fp8 | 970 | +4/-4 | 49,319 | Black Forest Labs | Open |
| 43 | dall-e-3 | 969 | +4/-4 | 240,462 | OpenAI | Proprietary |
| 44 | flux-1-kontext-dev | 942 | +4/-4 | 217,021 | Black Forest Labs | flux-1-dev-non-commercial-license |
| 45 | stable-diffusion-v35-large | 939 | +4/-4 | 23,379 | Stability AI | Open |
| 46 | bagel | 900 | +6/-6 | 12,455 | Bytedance | Apache 2.0 |
Data is for reference only. Official sources are authoritative. Click model names to view DataLearner model profiles.
本月 LMArena Text-to-Image Arena 排行榜更新至2026年2月12日,共收录46个模型,累计投票近392万次,是所有视觉创作类排行榜中模型最多、参与最广泛的赛道。
OpenAI GPT-Image-1.5 登顶。gpt-image-1.5-high-fidelity 以1248分位居榜首,超越了 Google 的 Gemini 系列,展示了 OpenAI 在文生图领域持续进化的能力。
Google Gemini 3 Pro 紧追。gemini-3-pro-image-preview-2k (nano-banana-pro)(1237分,第二)和 gemini-3-pro-image-preview (nano-banana-pro)(1233分,第三)展示了 Google 最新一代图像生成模型的实力。而 gemini-2.5-flash-image-preview(1157分,第八)凭借65万票的超高参与度证明了其广泛的用户基础。
xAI Grok 崛起。grok-imagine-image(1174分,第四)和 grok-imagine-image-pro(1166分,第六)首次活跃在文生图领域,均为初步评估阶段,但已展现出强劲的竞争力。
Black Forest Labs Flux 2 家族。flux-2-max(1169分,第五)、flux-2-flex(1158分,第七)、flux-2-pro(1156分,第九)和 flux-2-dev(1150分,第11)四个变体均稳居前列,Flux 2 系列在文生图领域确立了其作为顶级选择的地位。
国产模型在文生图领域展现了多样化的竞争力:
开源/可商用模型表现:
GPT-Image-1.5 High-Fidelity (OpenAI)
Elo 1248分,榜单冠军,OpenAI最新文生图旗舰
Gemini 3 Pro Image Preview 2K (Google)
Elo 1237分,Google第三代图像生成模型
Gemini 3 Pro Image Preview (Google)
Elo 1233分,与2K版本共同代表Google最强实力
HunyuanImage-3.0 (腾讯)
Elo 1151分,国产排名最高,位列全球第10
Seedream-4.5 (字节跳动)
Elo 1141分,位列全球第14,字节跳动新一代图像生成
Qwen-Image-2512 (阿里)
Elo 1139分,Apache 2.0开源,可企业商用
根据 LMArena Text-to-Image Arena 2026年2月排行榜 Elo 评分,GPT-Image-1.5 High-Fidelity(OpenAI,1248分)位居榜首,Gemini 3 Pro Image Preview 2K(Google,1237分)和 Gemini 3 Pro Image Preview(Google,1233分)分列亚军和季军。前三名均为闭源商业模型。
开源领域,Qwen-Image-2512(阿里,1139分,Apache 2.0)表现最为突出,位列全球第15。Z-Image-Turbo(阿里,1083分,Apache 2.0)和 GLM-Image(智谱,1013分,MIT)也是优秀的开源选择。值得注意的是,Bagel(字节跳动,900分,Apache 2.0)虽分数较低但提供了完全开放的方案。
国产模型整体表现强劲:HunyuanImage-3.0(腾讯,1151分,第10名)排名最高,Seedream-4.5(字节跳动,1141分)和 Qwen-Image-2512(阿里,1139分)紧随其后。阿里系模型在该领域布局最广,拥有5款以上入榜模型,涵盖文生图、图像编辑等多个方向。
FLUX 2 系列在本榜单中表现非常出色:Flux-2-Max(1169分,第5)、Flux-2-Flex(1158分,第7)、Flux-2-Pro(1156分,第9)和 Flux-2-Dev(1150分,第11)四个变体均位列全球前15。FLUX 2 系列是目前非大厂模型中表现最好的文生图方案。