Image Editing on KRIS-Bench
81.92Overall ScoreCritic-v2
Evaluation Results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Critic-v2Proprietary=false2026.05 | 81.92 | 82.09 | 84.79 | 76.7 | — | — | — | — | 85.76 | 89.5 | 58.44 | 84.4 | 84.91 | 77.46 | 75.94 | — | — | |
| Self-adaptive Interleaved ReasonerModel Type=Open-Source2026.05 | 80.18 | 84.24 | 74.83 | 85.53 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GPT-4oSource=Closed-Source2026.02 | 80.09 | 79.8 | 81.37 | 78.32 | 74.5 | 94.75 | 80.25 | 78.75 | 83.17 | 79.08 | 68.25 | 85.5 | 80.06 | 71.56 | 85.08 | — | — | |
| GPT-4oCategory=Proprietary Models2025.12 | 80.09 | 79.8 | 81.37 | 78.32 | — | — | — | — | 83.17 | 79.08 | 68.25 | 85.5 | 80.06 | 71.56 | 85.08 | — | — | |
| GPT-4oProprietary=true2026.05 | 80.09 | 79.8 | 81.37 | 78.32 | — | — | — | — | 83.17 | 79.08 | 68.25 | 85.5 | 80.06 | 71.56 | 85.08 | — | — | |
| GPT4oModel Type=Close-source2026.05 | 80.09 | 79.8 | 81.37 | 78.32 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-Image-Edit + EditThinker-Expert-GPT4.1Category=Think-while-Edit, Base Editor=Qwen-Image-Edit, Thinker Model=EditThinker-Expert-GPT4.1, Multi-image editing=not supported2025.12 | 79.34 | 81.67 | 80.91 | 73.4 | — | — | — | — | 83.7 | 76.08 | 0 | 81.99 | 80.53 | 71.94 | 76.07 | — | — | |
| Critic-v1Proprietary=false2026.05 | 79.02 | 79.96 | 82.76 | 71.15 | — | — | — | — | 84.06 | 86.42 | 55.97 | 83.35 | 82.57 | 67.91 | 74.39 | — | — | |
| DoubaoModel Type=Close-source2026.05 | 77.31 | 78.1 | 76.86 | 76.93 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Gemini 2.5 flashModel Type=Close-source2026.05 | 77.29 | 77.03 | 78.29 | 75.93 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FLUX.1 Kontext [Dev] + EditThinker-Expert-GPT4.1Category=Think-while-Edit, Base Editor=FLUX.1 Kontext [Dev], Thinker Model=EditThinker-Expert-GPT4.1, Multi-image editing=not supported2025.12 | 74.93 | 79.33 | 75.62 | 68.86 | — | — | — | — | 81.03 | 74.67 | 0 | 77.6 | 74.98 | 71.38 | 65.5 | — | — | |
| Emu3.5Model Type=Open-Source2026.05 | 73.75 | 78.59 | 71.92 | 71.14 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-Image-Edit + EditThinker-8BCategory=Think-while-Edit, Base Editor=Qwen-Image-Edit, Thinker Model=EditThinker-8B, Multi-image editing=not supported2025.12 | 71.91 | 77.24 | 72.02 | 65.94 | — | — | — | — | 78.48 | 73.83 | 0 | 76.2 | 70.69 | 65.23 | 66.89 | — | — | |
| UniComModel Category=Unified Multimodal Models2026.03 | 70.11 | 74.63 | 69.48 | 65.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FLUX.1 Kontext [Dev] + EditThinker-8BCategory=Think-while-Edit, Base Editor=FLUX.1 Kontext [Dev], Thinker Model=EditThinker-8B, Multi-image editing=not supported2025.12 | 69.53 | 73.73 | 70.09 | 63.6 | — | — | — | — | 77.82 | 65.5 | 0 | 73.44 | 69.04 | 62.29 | 65.33 | — | — | |
| Uni-CoTCategory=Open-source Models2025.12 | 68 | 71.85 | 67.16 | 63.68 | — | — | — | — | 72.76 | 72.87 | 67.1 | 70.81 | 66 | 53.43 | 73.93 | — | — | |
| Uni-CoTProprietary=false2026.05 | 68 | 71.85 | 67.16 | 63.68 | — | — | — | — | 72.76 | 72.87 | 67.1 | 70.81 | 66 | 53.43 | 73.93 | — | — | |
| WorldEditSource=Open-Source2026.02 | 66.86 | 56.98 | 68.17 | 56.98 | 76.18 | 85.91 | 53.73 | 39.8 | 71.94 | 71.5 | 76.13 | 64.35 | 69.39 | 50.92 | 65.06 | — | — | |
| Qwen-Image-EditCategory=Think-while-Edit, Test Source=Official re-test (‡), Multi-image editing=not supported2025.12 | 64.43 | 72.89 | 61.15 | 61.7 | — | — | — | — | 72.73 | 73.33 | 0 | 63.5 | 60.4 | 57.47 | 67.97 | — | — | |
| UNObase_model=BAGEL-think2026.05 | 62.57 | — | — | — | — | — | — | — | — | 72.25 | 59.18 | 64.65 | 61.51 | 48 | 55.67 | — | 73.09 | |
| Gemini 2.0Source=Closed-Source2026.02 | 62.41 | 65.26 | 59.65 | 62.9 | 69.5 | 81.75 | 47.75 | 53 | 66.33 | 63.33 | 63.92 | 68.19 | 56.94 | 54.13 | 71.67 | — | — | |
| Gemini 2.0Category=Proprietary Models2025.12 | 62.41 | 65.26 | 59.65 | 62.9 | — | — | — | — | 66.33 | 63.33 | 63.92 | 68.19 | 56.94 | 54.13 | 71.67 | — | — | |
| Gemini 2.0Proprietary=true2026.05 | 62.41 | 65.26 | 59.65 | 62.9 | — | — | — | — | 66.33 | 63.33 | 63.92 | 68.19 | 56.94 | 54.13 | 71.67 | — | — | |
| FLUX.1 Kontext [Dev]Category=Think-while-Edit, Test Source=Official re-test (‡), Multi-image editing=not supported2025.12 | 61.81 | 70.09 | 58.35 | 59.75 | — | — | — | — | 71.12 | 67.25 | 0 | 56.6 | 58.91 | 56.75 | 63.72 | — | — | |
| ReasonEdit-Q (thinking+reflection)Model Type=Open-Source2026.05 | 61.57 | 63.92 | 64.85 | 52.41 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Step 3o visionCategory=Proprietary Models2025.12 | 61.43 | 66.7 | 62.32 | 51.99 | — | — | — | — | 69.67 | 61.08 | 63.25 | 66.88 | 60.88 | 49.06 | 54.92 | — | — | |
| DoubaoSource=Closed-Source2026.02 | 60.7 | 63.3 | 62.23 | 54.17 | 66.75 | 89 | 57 | 48.75 | 70.92 | 59.17 | 40.58 | 65.5 | 61.19 | 47.75 | 60.58 | — | — | |
| DoubaoCategory=Proprietary Models2025.12 | 60.7 | 63.3 | 62.23 | 54.17 | — | — | — | — | 70.92 | 59.17 | 40.58 | 65.5 | 61.19 | 47.75 | 60.58 | — | — | |
| DoubaoProprietary=true2026.05 | 60.7 | 63.3 | 62.23 | 54.17 | — | — | — | — | 70.92 | 59.17 | 40.58 | 65.5 | 61.19 | 47.75 | 60.58 | — | — | |
| BAGEL-ThinkCategory=Open-source Models2025.12 | 60.18 | 66.18 | 61.92 | 49.02 | — | — | — | — | 67.42 | 68.33 | 58.67 | 63.55 | 61.4 | 48.12 | 50.22 | — | — | |
| BAGEL-think2026.05 | 60.18 | — | — | — | — | — | — | — | — | 68.33 | 58.67 | 63.55 | 61.4 | 48.12 | 50.22 | — | 67.42 | |
| BAGEL-ThinkProprietary=false2026.05 | 60.18 | 66.18 | 61.92 | 49.02 | — | — | — | — | 67.42 | 68.33 | 58.67 | 63.55 | 61.4 | 48.12 | 50.22 | — | — | |
| BAGEL-thinkingModel Type=Open-Source2026.05 | 60.18 | 66.18 | 61.92 | 49.02 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OmniGen2 + EditThinker-Expert-GPT4.1Category=Think-while-Edit, Base Editor=OmniGen2, Thinker Model=EditThinker-Expert-GPT4.1, Multi-image editing=not supported2025.12 | 57.34 | 63.22 | 57.22 | 51.26 | — | — | — | — | 65.55 | 56.83 | 0 | 63.35 | 55.26 | 44.38 | 60.44 | — | — | |
| BAGELCategory=Open-source Models2025.12 | 56.21 | 60.26 | 55.86 | 51.69 | — | — | — | — | 64.27 | 62.42 | 42.45 | 55.4 | 56.01 | 52.54 | 50.56 | — | — | |
| BAGELModel Category=Unified Multimodal Models2026.03 | 56.21 | 60.26 | 55.86 | 51.69 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BAGELModel Type=Open-Source2026.05 | 56.21 | 60.26 | 55.86 | 51.69 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-Image-Edit-2509Model Type=Open-Source2026.05 | 56.15 | 61.47 | 56.79 | 47.07 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniWorld-Qwen-Image-EditModel Type=Open-Source2026.05 | 55.98 | 61.72 | 56.38 | 46.69 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FLUX.1 Kontext [Pro]Model Category=Generation-only Models2026.03 | 54.17 | 57.22 | 55.06 | 46.69 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Bagel-ThinkSource=Open-Source2026.02 | 53.36 | 55.77 | 59.44 | 39.26 | 74.75 | 75 | 49.5 | 38.25 | 66.42 | 67.75 | 0 | 59.63 | 59.38 | 51.19 | 27.33 | — | — | |
| OmniGen2 + EditThinker-8BCategory=Think-while-Edit, Base Editor=OmniGen2, Thinker Model=EditThinker-8B, Multi-image editing=not supported2025.12 | 53.09 | 59.98 | 52.71 | 46.43 | — | — | — | — | 62.18 | 53.92 | 0 | 61.5 | 49.9 | 37.04 | 58.78 | — | — | |
| Step1X-Edit v1.1Category=Open-source Models, Multi-image editing=not supported2025.12 | 51.59 | 53.05 | 54.34 | 44.66 | — | — | — | — | 64.17 | 61.75 | 0 | 52.06 | 55.06 | 52.56 | 36.75 | — | — | |
| Step1X-Edit v1.1Model Type=Open-Source2026.05 | 51.59 | 53.05 | 54.34 | 44.66 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniWorld-FLUX.1-Kontext-DevModel Type=Open-Source2026.05 | 51.04 | 55.5 | 51.39 | 43.76 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OmniGen2Category=Think-while-Edit, Test Source=Official re-test (‡), Multi-image editing=not supported2025.12 | 50.52 | 58.73 | 48.42 | 45.84 | — | — | — | — | 60.21 | 54.67 | 0 | 53.6 | 46.76 | 37.67 | 56.67 | — | — | |
| UniWorld-V1Category=Open-source Models2025.12 | 50.27 | 47.71 | 44.8 | 47.92 | — | — | — | — | 58.17 | 54.5 | 63 | 47.5 | 43.94 | 42 | 53.83 | — | — | |
| Uniworld-V1Model Type=Open-Source2026.05 | 50.27 | 47.71 | 44.8 | 47.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OmniGen2Category=Open-source Models2025.12 | 49.71 | 57.36 | 44.2 | 47.79 | — | — | — | — | 59.92 | 52.25 | 54.75 | 47.56 | 43.12 | 32.5 | 63.08 | — | — | |
| OmniGen2Model Category=Unified Multimodal Models2026.03 | 49.71 | 57.36 | 44.2 | 47.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OmniGen2Model Type=Open-Source2026.05 | 49.71 | 57.36 | 44.2 | 47.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FLUX.1 Kontext [Dev]Category=Open-source Models, Multi-image editing=not supported2025.12 | 49.54 | 53.28 | 50.36 | 42.53 | — | — | — | — | 64.83 | 60.92 | 0 | 48.94 | 50.81 | 46.06 | 39 | — | — | |
| Flux-Kontext-devModel Type=Open-Source2026.05 | 49.54 | 53.28 | 50.36 | 42.53 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BagelSource=Open-Source2026.02 | 47.76 | 47.71 | 52.17 | 40.23 | 66.75 | 67 | 40.5 | 25.25 | 58.08 | 54.5 | 0 | 52.69 | 52 | 49.63 | 30.83 | — | — | |
| ByteMorphCategory=Open-source Models, Multi-image editing=not supported2025.12 | 44.85 | 51.27 | 46.92 | 31.67 | — | — | — | — | 61.17 | 62 | 0 | 45.5 | 47.38 | 32 | 31.33 | — | — | |
| HiDream-E1Category=Open-source Models, Multi-image editing=not supported2025.12 | 44.72 | 43.31 | 50.05 | 37.64 | — | — | — | — | 52.75 | 49.42 | 0 | 52.56 | 49.25 | 45.19 | 30.08 | — | — | |
| Step1X-EditSource=Open-Source2026.02 | 43.29 | 45.52 | 48.01 | 31.82 | 63 | 70.25 | 33.25 | 17.5 | 55.5 | 51.75 | 0 | 44.69 | 49.06 | 40.88 | 22.75 | — | — | |
| Step1X-EditCategory=Open-source Models, Multi-image editing=not supported2025.12 | 43.29 | 45.52 | 48.01 | 31.82 | — | — | — | — | 55.5 | 51.75 | 0 | 44.69 | 49.06 | 40.88 | 22.75 | — | — | |
| Step1X-EditModel Category=Specialized Editing Models2026.03 | 43.29 | 45.52 | 48.01 | 31.82 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Emu2Source=Open-Source2026.02 | 39.7 | 45.4 | 37.54 | 34.91 | 47.75 | 75.25 | 31.5 | 11.25 | 51.5 | 48.83 | 22.17 | 34.69 | 38.44 | 24.81 | 45 | — | — | |
| Emu2Category=Open-source Models2025.12 | 39.7 | 45.4 | 37.54 | 34.91 | — | — | — | — | 51.5 | 48.83 | 22.17 | 34.69 | 38.44 | 24.81 | 45 | — | — | |
| AnyeditSource=Open-Source2026.02 | 38.55 | 39.26 | 41.88 | 31.74 | 54.75 | 67.5 | 20.75 | 10.5 | 47.67 | 45.17 | 0 | 38.56 | 42.94 | 36.56 | 26.92 | — | — | |
| AnyEditCategory=Open-source Models, Multi-image editing=not supported2025.12 | 38.55 | 39.26 | 41.88 | 31.74 | — | — | — | — | 47.67 | 45.17 | 0 | 38.56 | 42.94 | 36.56 | 26.92 | — | — | |
| AnyEditModel Category=Specialized Editing Models2026.03 | 38.55 | 39.26 | 41.88 | 31.74 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MagicbrushSource=Open-Source2026.02 | 37.15 | 41.84 | 39.24 | 26.54 | 53.5 | 76.25 | 32 | 20.5 | 53.92 | 39.58 | 0 | 42.94 | 38.06 | 30 | 23.08 | — | — | |
| MagicBrushCategory=Open-source Models, Multi-image editing=not supported2025.12 | 37.15 | 41.84 | 39.24 | 26.54 | — | — | — | — | 53.92 | 39.58 | 0 | 42.94 | 38.06 | 30 | 23.08 | — | — | |
| MagicBrushModel Category=Specialized Editing Models2026.03 | 37.15 | 41.84 | 39.24 | 26.54 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OmnigenSource=Open-Source2026.02 | 28.85 | 33.11 | 28.02 | 23.89 | 35.75 | 49.5 | 28.5 | 16.75 | 37.92 | 28.25 | 21.83 | 30.63 | 27.19 | 11.94 | 35.83 | — | — | |
| OmniGenCategory=Open-source Models2025.12 | 28.85 | 33.11 | 28.02 | 23.89 | — | — | — | — | 37.92 | 28.25 | 21.83 | 30.63 | 27.19 | 11.94 | 35.83 | — | — | |
| OmniGenModel Category=Unified Multimodal Models2026.03 | 28.85 | 33.11 | 28.02 | 23.89 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ip2pSource=Open-Source2026.02 | 22.82 | 23.33 | 25.59 | 17.28 | 17.5 | 55.5 | 18 | 10.25 | 30.33 | 21.33 | 0 | 22.56 | 26.56 | 19.81 | 14.75 | — | — | |
| InstructPix2PixCategory=Open-source Models, Multi-image editing=not supported2025.12 | 22.82 | 23.33 | 25.59 | 17.28 | — | — | — | — | 30.33 | 21.33 | 0 | 22.56 | 26.56 | 19.81 | 14.75 | — | — | |
| Instruct-Pix2PixModel Category=Specialized Editing Models2026.03 | 22.82 | 23.33 | 25.59 | 17.28 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GPT4o (Apr. 2025)Release Date=Apr. 20252025.11 | 0.8009 | 0.798 | 0.8137 | 0.7832 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GPT4o (Sep. 2025)Release Date=Sep. 20252025.11 | 0.79 | 0.8116 | 0.7824 | 0.7709 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Doubao (Seedream 4.0)Version=Seedream 4.0, Release Date=Aug. 20252025.11 | 0.7731 | 0.781 | 0.7686 | 0.7693 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Gemini 2.5 flash (Sep. 2025)Release Date=Sep. 20252025.11 | 0.7729 | 0.7703 | 0.7829 | 0.7593 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Gemini 2 flash (Apr. 2025)Release Date=Apr. 20252025.11 | 0.6241 | 0.6526 | 0.5965 | 0.629 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReasonEdit-Q (thinking+reflection)thinking=true, reflection=true2025.11 | 0.6157 | 0.6392 | 0.6485 | 0.5241 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReasonEdit-S (thinking+reflection)thinking=true, reflection=true2025.11 | 0.6093 | 0.6244 | 0.6572 | 0.5042 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReasonEdit-Q (thinking)thinking=true, reflection=false2025.11 | 0.6081 | 0.6244 | 0.6449 | 0.5202 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Doubao (seed edit 1.6)Version=seed edit 1.6, Release Date=Apr. 20252025.11 | 0.607 | 0.633 | 0.6223 | 0.5417 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BAGEL-thinkingthinking=true2025.11 | 0.6018 | 0.6618 | 0.6192 | 0.4902 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReasonEdit-S (thinking)thinking=true, reflection=false2025.11 | 0.5864 | 0.5979 | 0.6276 | 0.4978 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReasonEdit-Q (base)thinking=false, reflection=false2025.11 | 0.5805 | 0.6229 | 0.6222 | 0.4453 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReasonEdit-S (base)thinking=false, reflection=false2025.11 | 0.5633 | 0.5823 | 0.6055 | 0.4621 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BAGELthinking=false2025.11 | 0.5621 | 0.6026 | 0.5586 | 0.5169 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Qwen-Image-Edit-25092025.11 | 0.5615 | 0.6147 | 0.5679 | 0.4707 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniWorld-Qwen-Image-Edit2025.11 | 0.5598 | 0.6172 | 0.5638 | 0.4669 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Step1X-Edit v1.12025.11 | 0.5159 | 0.5305 | 0.5434 | 0.4466 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniWorld-FLUX.1-Kontext-Dev2025.11 | 0.5104 | 0.555 | 0.5139 | 0.4376 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Uniworld-V12025.11 | 0.5027 | 0.4771 | 0.448 | 0.4792 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Omnigen 22025.11 | 0.4971 | 0.5736 | 0.442 | 0.4779 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Flux-Kontext-dev2025.11 | 0.4954 | 0.5328 | 0.5036 | 0.4253 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Hidream-I1 (E1)2025.11 | 0.4472 | 0.4331 | 0.5005 | 0.3764 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Hidream-E1.12025.11 | 0.4225 | 0.4352 | 0.4471 | 0.3608 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ICEdit2025.11 | 0.407 | 0.4699 | 0.4273 | 0.2776 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Omnigen2025.11 | 0.2885 | 0.3311 | 0.2802 | 0.2389 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AnyEdit2025.12 | — | — | — | — | 62 | 66.75 | 15 | 10.5 | — | — | — | — | — | — | — | 38.36 | — | |
| AnyEdit2025.12 | — | — | — | — | 68.25 | 84.5 | 12 | 11 | — | — | — | — | — | — | — | 43.94 | — |