Instructional Image Editing on CV-Arena 12K examples 1.0
1,187Elo RatingQwen Image Edit
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen Image EditEvaluation Protocol=Human Only2026.05 | 1,187 | 31 | |
| nano bananaEvaluation Protocol=EditReward Only2026.05 | 1,186 | 31 | |
| SuperEditEvaluation Protocol=Human Only2026.05 | 1,169 | 46 | |
| nano banana proEvaluation Protocol=Human Only2026.05 | 1,151 | 75 | |
| CV-AgentEvaluation Protocol=Active Elo (Ours)2026.05 | 1,145 | 50 | |
| nano banana proEvaluation Protocol=EditReward Only2026.05 | 1,141 | 35 | |
| UniWorld v1Evaluation Protocol=Human Only2026.05 | 1,133 | 30 | |
| Step1X EditEvaluation Protocol=EditReward Only2026.05 | 1,129 | 28 | |
| nano banana proEvaluation Protocol=Active Elo (Ours)2026.05 | 1,127 | 48 | |
| CV-AgentEvaluation Protocol=Human Only2026.05 | 1,116 | 82 | |
| ManusEvaluation Protocol=Active Elo (Ours)2026.05 | 1,109 | 26 | |
| CV-AgentEvaluation Protocol=EditReward Only2026.05 | 1,102 | 33 | |
| VAREditEvaluation Protocol=Human Only2026.05 | 1,098 | 43 | |
| Step1X EditEvaluation Protocol=CV-Judge only2026.05 | 1,095 | 24 | |
| gpt-image-1.5Evaluation Protocol=Active Elo (Ours)2026.05 | 1,091 | 42 | |
| ManusEvaluation Protocol=Human Only2026.05 | 1,080 | 75 | |
| ICEditEvaluation Protocol=CV-Judge only2026.05 | 1,075 | 35 | |
| seeddream4.5Evaluation Protocol=Active Elo (Ours)2026.05 | 1,073 | 39 | |
| MagicBrushEvaluation Protocol=EditReward Only2026.05 | 1,067 | 28 | |
| ByteMorphEvaluation Protocol=Human Only2026.05 | 1,062 | 49 | |
| nano bananaEvaluation Protocol=Active Elo (Ours)2026.05 | 1,056 | 40 | |
| gpt-image-1.5Evaluation Protocol=CV-Judge only2026.05 | 1,054 | 24 | |
| MagicBrushEvaluation Protocol=Human Only2026.05 | 1,044 | 39 | |
| VAREditEvaluation Protocol=EditReward Only2026.05 | 1,043 | 28 | |
| gpt-image-1Evaluation Protocol=EditReward Only2026.05 | 1,039 | 38 | |
| gpt-image-1Evaluation Protocol=Active Elo (Ours)2026.05 | 1,038 | 31 | |
| nano banana proEvaluation Protocol=CV-Judge only2026.05 | 1,034 | 38 | |
| Instruct-CLIPEvaluation Protocol=Human Only2026.05 | 1,027 | 58 | |
| ManusEvaluation Protocol=EditReward Only2026.05 | 1,022 | 31 | |
| Flux2Evaluation Protocol=Active Elo (Ours)2026.05 | 1,020 | 23 | |
| Flux2Evaluation Protocol=CV-Judge only2026.05 | 1,014 | 26 | |
| UniWorld v1Evaluation Protocol=EditReward Only2026.05 | 1,010 | 41 | |
| Step1X EditEvaluation Protocol=Human Only2026.05 | 1,009 | 59 | |
| Edit R1Evaluation Protocol=Active Elo (Ours)2026.05 | 1,002 | 49 | |
| CV-AgentEvaluation Protocol=CV-Judge only2026.05 | 993 | 37 | |
| JarvisEvoEvaluation Protocol=Human Only2026.05 | 991 | 68 | |
| wan 2.5 previewEvaluation Protocol=Active Elo (Ours)2026.05 | 984 | 36 | |
| JarvisEvoEvaluation Protocol=EditReward Only2026.05 | 984 | 29 | |
| gpt-image-1.5Evaluation Protocol=Human Only2026.05 | 974 | 45 | |
| seeddream4.5Evaluation Protocol=CV-Judge only2026.05 | 973 | 31 | |
| AnysdEvaluation Protocol=EditReward Only2026.05 | 969 | 29 | |
| Qwen Image EditEvaluation Protocol=Active Elo (Ours)2026.05 | 965 | 36 | |
| ICEditEvaluation Protocol=Human Only2026.05 | 956 | 42 | |
| Edit R1Evaluation Protocol=CV-Judge only2026.05 | 953 | 28 | |
| gpt-image-1.5Evaluation Protocol=EditReward Only2026.05 | 952 | 36 | |
| VAREditEvaluation Protocol=Active Elo (Ours)2026.05 | 948 | 33 | |
| Flux2Evaluation Protocol=Human Only2026.05 | 938 | 29 | |
| Step1X EditEvaluation Protocol=Active Elo (Ours)2026.05 | 932 | 30 | |
| wan 2.5 previewEvaluation Protocol=CV-Judge only2026.05 | 932 | 30 | |
| Edit R1Evaluation Protocol=EditReward Only2026.05 | 931 | 38 | |
| seeddream4.5Evaluation Protocol=Human Only2026.05 | 920 | 70 | |
| UniWorld v1Evaluation Protocol=Active Elo (Ours)2026.05 | 915 | 38 | |
| gpt-image-1Evaluation Protocol=CV-Judge only2026.05 | 912 | 41 | |
| Instruct-CLIPEvaluation Protocol=EditReward Only2026.05 | 909 | 29 | |
| Edit R1Evaluation Protocol=Human Only2026.05 | 902 | 75 | |
| MagicBrushEvaluation Protocol=Active Elo (Ours)2026.05 | 897 | 31 | |
| SuperEditEvaluation Protocol=EditReward Only2026.05 | 894 | 25 | |
| nano bananaEvaluation Protocol=CV-Judge only2026.05 | 892 | 35 | |
| gpt-image-1Evaluation Protocol=Human Only2026.05 | 884 | 78 | |
| SuperEditEvaluation Protocol=Active Elo (Ours)2026.05 | 880 | 37 | |
| ManusEvaluation Protocol=CV-Judge only2026.05 | 874 | 34 | |
| seeddream4.5Evaluation Protocol=EditReward Only2026.05 | 873 | 32 | |
| nano bananaEvaluation Protocol=Human Only2026.05 | 866 | 74 | |
| ByteMorphEvaluation Protocol=Active Elo (Ours)2026.05 | 864 | 30 | |
| SuperEditEvaluation Protocol=CV-Judge only2026.05 | 856 | 27 | |
| AnysdEvaluation Protocol=Human Only2026.05 | 848 | 64 | |
| ICEditEvaluation Protocol=Active Elo (Ours)2026.05 | 847 | 41 | |
| ByteMorphEvaluation Protocol=EditReward Only2026.05 | 845 | 30 | |
| JarvisEvoEvaluation Protocol=CV-Judge only2026.05 | 839 | 42 | |
| Instruct-CLIPEvaluation Protocol=Active Elo (Ours)2026.05 | 831 | 19 | |
| wan 2.5 previewEvaluation Protocol=Human Only2026.05 | 830 | 43 | |
| Flux2Evaluation Protocol=EditReward Only2026.05 | 823 | 37 | |
| Qwen Image EditEvaluation Protocol=CV-Judge only2026.05 | 821 | 21 | |
| AnysdEvaluation Protocol=Active Elo (Ours)2026.05 | 815 | 35 | |
| VAREditEvaluation Protocol=CV-Judge only2026.05 | 804 | 18 | |
| JarvisEvoEvaluation Protocol=Active Elo (Ours)2026.05 | 799 | 32 | |
| ICEditEvaluation Protocol=EditReward Only2026.05 | 794 | 39 | |
| AnysdEvaluation Protocol=CV-Judge only2026.05 | 787 | 37 | |
| Qwen Image EditEvaluation Protocol=EditReward Only2026.05 | 771 | 30 | |
| UniWorld v1Evaluation Protocol=CV-Judge only2026.05 | 770 | 26 | |
| wan 2.5 previewEvaluation Protocol=EditReward Only2026.05 | 756 | 35 | |
| MagicBrushEvaluation Protocol=CV-Judge only2026.05 | 754 | 25 | |
| ByteMorphEvaluation Protocol=CV-Judge only2026.05 | 738 | 32 | |
| Instruct-CLIPEvaluation Protocol=CV-Judge only2026.05 | 722 | 36 |