Structured Image Editing on StructEditBench
64.64Math AccuracyNano Banana 2.0
Evaluation Results
| Method | Links | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Nano Banana 2.02025.10 | 64.64 | 24.88 | — | 63.28 | 16.74 | — | 72.58 | 26.39 | — | 81.07 | 28.12 | — | 71.19 | 26.71 | — | 91.1 | 25.14 | — | 67.05 | 25.59 | — | |
| Ours2025.10 | 54.74 | 23.31 | — | 50.58 | 15.33 | — | 60.18 | 24.65 | — | 73 | 26.33 | — | 75.05 | 25.8 | — | 77.08 | 23.19 | — | 55.98 | 24.01 | — | |
| GPT-Image2025.10 | 51.49 | 17.06 | — | 45.82 | 12.56 | — | 50.71 | 17.24 | — | 76.03 | 16.55 | — | 67.61 | 16.61 | — | 83.26 | 14.35 | — | 52.2 | 16.64 | — | |
| Seedream 4.02025.10 | 51.06 | 23.63 | — | 46.83 | 16.54 | — | 51.72 | 24.12 | — | 71.13 | 26.93 | — | 69.22 | 26.46 | — | 88.19 | 24.75 | — | 52.85 | 24.45 | — | |
| Nano Banana2025.10 | 50.46 | 20.77 | — | 46.42 | 14.93 | — | 52.97 | 21.22 | — | 66.56 | 22.92 | — | 69.16 | 22.61 | — | 75.71 | 19.75 | — | 51.57 | 21.09 | — | |
| Seedream 4.0Evaluator=Qwen2.5-VL-72B2025.10 | 41.31 | 23.63 | 95.87 | 43.17 | 16.54 | 79.92 | 41.18 | 24.12 | 94.33 | 59.94 | 26.93 | 93.84 | 53.39 | 26.46 | 95.18 | 78.56 | 24.75 | 93.83 | 46.02 | 24.45 | 95.14 | |
| OursEvaluator=Qwen2.5-VL-72B2025.10 | 41.1 | 23.31 | 95.14 | 40.44 | 15.33 | 75.28 | 37.2 | 24.65 | 96.26 | 57.58 | 26.33 | 96.47 | 52.46 | 25.8 | 96.38 | 69.75 | 23.19 | 94.38 | 43.56 | 24.01 | 85.35 | |
| Qwen-Edit2025.10 | 40.48 | 23.73 | — | 30.17 | 12.33 | — | 44.83 | 26.11 | — | 53.74 | 27.31 | — | 55.99 | 25.53 | — | 67.76 | 25.71 | — | 38.12 | 24.81 | — | |
| GPT-ImageEvaluator=Qwen2.5-VL-72B2025.10 | 39.77 | 17.06 | 91.99 | 39.61 | 12.56 | 65.8 | 47.77 | 17.24 | 93.57 | 59.01 | 16.55 | 91.39 | 53.23 | 16.61 | 92.93 | 78.02 | 14.35 | 84 | 44 | 16.64 | 91.19 | |
| Nano BananaEvaluator=Qwen2.5-VL-72B2025.10 | 39.46 | 20.77 | 95.26 | 42.33 | 14.93 | 73.92 | 46.72 | 21.22 | 96.16 | 54.35 | 22.92 | 96.21 | 53.33 | 22.61 | 96.46 | 66.87 | 19.75 | 92.6 | 44.46 | 21.09 | 95.24 | |
| Bagel-Think2025.10 | 37.4 | 23.97 | — | 28.98 | 16.82 | — | 42.51 | 26.49 | — | 36.11 | 26.75 | — | 43.15 | 25.57 | — | 40.46 | 23.83 | — | 33.34 | 24.7 | — | |
| FLUX.1 Kontext2025.10 | 37.36 | 19.78 | — | 32.29 | 14.61 | — | 39.12 | 20.1 | — | 58.35 | 20.38 | — | 50.39 | 20.99 | — | 58.05 | 18.52 | — | 37.56 | 19.84 | — | |
| Step1X-Edit2025.10 | 34.47 | 23.41 | — | 28.05 | 16.68 | — | 33.26 | 24.56 | — | 60.48 | 25.94 | — | 46.47 | 24.98 | — | 57.81 | 23.97 | — | 34.11 | 24.03 | — | |
| Qwen-EditEvaluator=Qwen2.5-VL-72B2025.10 | 34.03 | 23.73 | 97.14 | 25.5 | 12.33 | 65.93 | 38.23 | 26.11 | 98.14 | 43.25 | 27.31 | 97.94 | 43.08 | 25.53 | 97.42 | 58.82 | 25.71 | 96.65 | 31.99 | 24.81 | 97.31 | |
| Ovis-U12025.10 | 31.64 | 18.45 | — | 21.94 | 13.3 | — | 38.03 | 19.01 | — | 42.08 | 17.92 | — | 44.52 | 18.68 | — | 35.58 | 16.62 | — | 28.06 | 18.25 | — | |
| Flux KontextEvaluator=Qwen2.5-VL-72B2025.10 | 29.78 | 19.78 | 95.12 | 27.78 | 14.61 | 73.24 | 30.25 | 20.1 | 95.54 | 46.32 | 20.38 | 95.01 | 35.6 | 20.99 | 96.27 | 52.22 | 18.52 | 91.27 | 31.13 | 19.84 | 94.78 | |
| OmniGen22025.10 | 29.44 | 15.95 | — | 18.55 | 12.44 | — | 34.63 | 11.31 | — | 28.61 | 16.51 | — | 39.55 | 15.6 | — | 30.36 | 16.6 | — | 24.3 | 15.49 | — | |
| HiDream-E1.12025.10 | 28.07 | 18.43 | — | 26.36 | 12.91 | — | 29.63 | 18.26 | — | 43.77 | 18.04 | — | 36.66 | 16.47 | — | 48.79 | 17.12 | — | 29.63 | 18.01 | — | |
| Bagel-ThinkEvaluator=Qwen2.5-VL-72B2025.10 | 27.82 | 23.97 | 97.34 | 23.38 | 16.82 | 82.27 | 32.3 | 26.49 | 98.23 | 23.22 | 26.75 | 98.09 | 33.19 | 25.57 | 98.04 | 35.73 | 23.83 | 97.31 | 26.17 | 24.7 | 97.59 | |
| Step1X-EditEvaluator=Qwen2.5-VL-72B2025.10 | 27.22 | 23.41 | 97.11 | 24.99 | 16.68 | 81.54 | 25.11 | 24.56 | 97.75 | 43.95 | 25.94 | 97.86 | 33.49 | 24.98 | 97.38 | 48.48 | 23.97 | 96.87 | 28.26 | 24.03 | 97.26 | |
| DiMOO2025.10 | 26.79 | 21.56 | — | 16.52 | 14.98 | — | 24.03 | 21.77 | — | 29.52 | 22.26 | — | 26.08 | 22.57 | — | 24.64 | 19.47 | — | 21 | 21.49 | — | |
| Ovis-U1Evaluator=Qwen2.5-VL-72B2025.10 | 25.57 | 18.45 | 93.56 | 13.48 | 13.3 | 68.54 | 28.42 | 19.01 | 95.01 | 24.59 | 17.92 | 92.4 | 22 | 18.68 | 94.55 | 33.03 | 16.62 | 87.96 | 19.4 | 18.25 | 93 | |
| OmniGen2Evaluator=Qwen2.5-VL-72B2025.10 | 23.65 | 15.95 | 82.06 | 13.44 | 12.44 | 71.87 | 23.68 | 11.31 | 78.34 | 16.56 | 16.51 | 86.54 | 26.02 | 15.6 | 86.15 | 27.52 | 16.6 | 88.88 | 18.1 | 15.49 | 83.34 | |
| DiMOOEvaluator=Qwen2.5-VL-72B2025.10 | 22.22 | 21.56 | 95.19 | 8.78 | 14.98 | 74.37 | 16.3 | 21.77 | 96 | 21.18 | 22.26 | 95.68 | 18.11 | 22.57 | 96.26 | 20.6 | 19.47 | 92.09 | 14.22 | 21.49 | 95.04 | |
| Hidream-E1.1Evaluator=Qwen2.5-VL-72B2025.10 | 21.97 | 18.43 | 93.99 | 20.25 | 12.91 | 70.59 | 22.4 | 18.26 | 94.66 | 29.16 | 18.04 | 93.89 | 25.42 | 16.47 | 93.77 | 46.83 | 17.12 | 91.18 | 23.07 | 18.01 | 93.68 | |
| Bagel2025.10 | 21.27 | 21.38 | — | 27.11 | 16.38 | — | 29.94 | 22.7 | — | 41.59 | 24.22 | — | 47.16 | 23.56 | — | 47.35 | 21.54 | — | 28.87 | 22.06 | — | |
| BagelEvaluator=Qwen2.5-VL-72B2025.10 | 16.07 | 21.38 | 96.03 | 24.27 | 16.38 | 81.78 | 20.32 | 22.7 | 97.27 | 35.44 | 24.22 | 97.28 | 32.58 | 23.56 | 97.36 | 45.37 | 21.54 | 95.42 | 24.24 | 22.06 | 96.36 | |
| UniworldEvaluator=Qwen2.5-VL-72B2025.10 | 15.97 | 8.84 | 82.64 | 6.19 | 7.87 | 48.83 | 10.27 | 6.16 | 72.72 | 9.98 | 7.71 | 76.19 | 16.01 | 7.67 | 76.72 | 15.08 | 8.24 | 72.97 | 9.98 | 8.21 | 78.95 | |
| UniWorld-V12025.10 | 9.41 | 8.84 | — | 5.99 | 7.87 | — | 8.83 | 6.16 | — | 9.11 | 7.71 | — | 19.91 | 7.67 | — | 16.13 | 8.24 | — | 8.4 | 8.21 | — |