Evaluation of Image Generation Capabilities of Artificial Intelligence Models / 人工智能模型图像生成能力综合评测
by Zhenhui (Jack) Jiang1 , Zhengyu Wu1 , Jiaxin Li1 , Haozhe Xu2 , Yifan Wu1 ,Yi Lu1 / 蒋镇辉1 ，武正昱1 ，李佳欣1 ，徐昊哲2 ，吴轶凡1 ，鲁艺1 1 HKU Business School, 2 School of Management, Xi'an Jiaotong Universityjiangz@hku.hk .
🎨 New Img Generation
🖼️ Img Revision
Select a Leaderboard
Option 1: New Image Generation Quality Ranking
----Dimension 1-Alignment with Instruction
----Dimension 2-Image Integrity
----Dimension 3-Image Aesthetics/option>
Option 2: Safety and Responsibility Ranking
Rank Model Elo 1 Dreamina (即梦AI) 1123 2 ERNIE Bot (文心一言) V3.2.0 1105 3 Midjourney v6.1 1094 4 DouBao (豆包) 1084 5 MiaoBiShengHua (妙笔生画) 1083 6 FLUX.1 Pro 1079 7 GPT-4o 1058 8 Gemini 1.5 Pro 1045 9 DALL-E3 1025 10 SenseChat (商量) -5 1022 11 SenseMirage (秒画) V5.0 1014 12 Hunyuan-DiT（混元生图） 1005 12 Playground v2.5 1005 14 Imagen 3 1000 15 Stable Diffusion 3 Large 995 16 Spark (讯飞星火) 969 17 CogView3 - Plus 953 17 Qwen (通义千问) V2.5.0 953 19 WenXinYiGe (文心一格2) 890 20 TongYiWanXiang (通义万相)wanx-v2 854 21 360 ZhiHui (360智绘) 834 22 DeepSeek Janus-Pro 810
Rank Model Elo 1 ERNIE Bot (文心一言) V3.2.0 1092 2 DouBao (豆包) 1087 3 MiaoBiShengHua (妙笔生画) 1074 3 Midjourney v6.1 1074 5 Dreamina (即梦AI) 1073 6 GPT-4o 1070 7 DALL-E3 1066 8 Gemini 1.5 Pro 1058 9 FLUX.1 Pro 1041 10 SenseChat (商量) -5 1005 11 Hunyuan-DiT（混元生图） 999 11 SenseMirage (秒画) V5.0 999 13 Stable Diffusion 3 Large 997 14 Spark (讯飞星火) 990 15 Imagen 3 986 16 CogView3 - Plus 980 17 Qwen (通义千问) V2.5.0 979 18 Playground v2.5 959 19 DeepSeek Janus-Pro 935 20 TongYiWanXiang (通义万相)wanx-v2 869 21 WenXinYiGe (文心一格2) 863 22 360 ZhiHui(360智绘) 804
Rank Model Elo 1 Dreamina (即梦AI) 1116 2 FLUX.1 Pro 1097 3 ERNIE Bot (文心一言) V3.2.0 1087 4 Midjourney v6.1 1078 5 妙笔生画 1066 6 DouBao (豆包) 1062 6 Gemini 1.5 Pro 1062 8 GPT-4o 1045 9 Imagen 3 1023 10 Hunyuan-DiT（混元生图） 1018 11 DALL-E3 1014 12 SenseChat (商量) -5 1012 13 Playground v2.5 1007 14 SenseMirage (秒画) V5.0 994 15 Stable Diffusion 3 Large 984 16 Spark (讯飞星火) 981 17 Qwen (通义千问) V2.5.0 943 18 CogView3 - Plus 931 19 360 ZhiHui(360智绘) 909 20 WenXinYiGe (文心一格2) 895 21 TongYiWanXiang (通义万相)wanx-v2 865 22 DeepSeek Janus-Pro 814
Rank Model Elo 1 Dreamina (即梦AI) 1141 2 ERNIE Bot (文心一言) V3.2.0 1084 3 Midjourney v6.1 1082 4 SenseChat (商量) -5 1065 5 DouBao (豆包) 1061 6 FLUX.1 Pro 1058 7 妙笔生画 1056 8 GPT-4o 1045 9 SenseMirage (秒画) V5.0 1039 10 Playground v2.5 1028 11 DALL-E3 1014 12 Gemini 1.5 Pro 1007 13 Hunyuan-DiT（混元生图） 995 14 Imagen 3 993 15 Stable Diffusion 3 Large 980 16 WenXinYiGe (文心一格2) 964 17 Qwen (通义千问) V2.5.0 960 18 Spark (讯飞星火) 955 19 CogView3 - Plus 939 20 TongYiWanXiang (通义万相)wanx-v2 869 21 360 ZhiHui(360智绘) 858 22 DeepSeek Janus-Pro 803
Rank Model Score 1 GPT-4o 6.04 2 Qwen (通义千问) V2.5.0 5.49 3 Gemini 1.5 Pro 5.23 4 Spark (讯飞星火) 4.44 5 Hunyuan-DiT（混元生图） 4.42 6 360 ZhiHui(360智绘) 4.27 7 Imagen 3 4.1 8 SenseChat (商量) -5 4.05 9 DouBao (豆包) 4.03 10 FLUX.1 Pro 3.94 11 SenseMirage (秒画) V5.0 3.88 12 DALL-E3 3.51 13 MiaoBiShengHua (妙笔生画) 3.47 14 ERNIE Bot (文心一言) V3.2.0 3.35 15 TongYiWanXiang (通义万相)wanx-v2 3.26 16 WenXinYiGe (文心一格2) 3.22 17 CogView3 - Plus 2.86 18 Dreamina (即梦AI) 2.63 19 Stable Diffusion 3 Large 2.35 20 Midjourney v6.1 2.29 21 DeepSeek Janus-Pro 2.19 22 Playground v2.5 1.79
Select a Leaderboard
Image Revision Test Ranking
----Dimension 1-Alignment with Reference
----Dimension 2-Revised Image Integrity
----Dimension 3-Revised Image Aesthetics
Rank Model Score 1 DouBao (豆包) 5.3 2 Dreamina (即梦AI) 5.2 3 ERNIE Bot (文心一言) V3.2.0 5.16 4 GPT-4o 5.02 5 Gemini 1.5 Pro 4.97 6 MiaoBiShengHua (妙笔生画) 4.71 7 Midjourney v6.1 4.66 7 SenseMirage (秒画) V5.0 4.66 9 CogView3 - Plus 4.58 10 Qwen (通义千问) V2.5.0 4.39 11 TongYiWanXiang (通义万相)wanx-v2 4.25 12 360 ZhiHui(360智绘) 3.85 13 WenXinYiGe (文心一格2) 3.05
Rank Model Score 1 Dreamina (即梦AI) 6.03 2 DouBao (豆包) 4.97 3 MiaoBiShengHua (妙笔生画) 4.94 4 ERNIE Bot (文心一言) V3.2.0 4.93 5 GPT-4o 4.92 6 SenseMirage (秒画) V5.0 4.78 7 CogView3 - Plus 4.6 8 TongYiWanXiang (通义万相)wanx-v2 4.55 9 Gemini 1.5 Pro 4.29 10 Qwen (通义千问) V2.5.0 4.22 10 360 ZhiHui(360智绘) 4.22 12 Midjourney v6.1 4.19 13 WenXinYiGe (文心一格2) 3.23
Rank Model Score 1 ERNIE Bot (文心一言) V3.2.0 5.43 2 DouBao (豆包) 5.4 2 Gemini 1.5 Pro 5.4 4 GPT-4o 5.12 5 Dreamina (即梦AI) 4.75 6 CogView3 - Plus 4.64 7 MiaoBiShengHua (妙笔生画) 4.55 8 Qwen (通义千问) V2.5.0 4.54 9 Midjourney v6.1 4.5 10 SenseMirage (秒画) V5.0 4.43 11 TongYiWanXiang (通义万相)wanx-v2 4.06 12 360 ZhiHui(360智绘) 3.52 13 WenXinYiGe (文心一格2) 3.15
Rank Model Score 1 DouBao (豆包) 5.53 2 Midjourney v6.1 5.29 3 Gemini 1.5 Pro 5.23 4 ERNIE Bot (文心一言) V3.2.0 5.12 5 GPT-4o 5.03 6 Dreamina (即梦AI) 4.83 7 SenseMirage (秒画) V5.0 4.75 8 MiaoBiShengHua (妙笔生画) 4.64 9 CogView3 - Plus 4.49 10 Qwen (通义千问) V2.5.0 4.42 11 TongYiWanXiang (通义万相)wanx-v2 4.14 12 360 ZhiHui(360智绘) 3.83 13 WenXinYiGe (文心一格2) 2.76
