Image Editing on IntelligentBench
78.9ScoreGPT-4o
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4oType=Private, Evaluation Protocol=Subset of answered cases only2025.05 | 78.9 | |
| GPT-4oType=Private2026.03 | 78.9 | |
| LVRPOType=Open-source2026.03 | 58.7 | |
| Gemini 2.0Type=Private, Evaluation Protocol=Subset of answered cases only2025.05 | 57.6 | |
| Gemini 2.0Type=Private2026.03 | 57.6 | |
| BAGELType=Open-source, Self-CoT=true2025.05 | 55.3 | |
| BAGELType=Open-source2026.03 | 55.3 | |
| BAGELType=Open-source, Self-CoT=false2025.05 | 44.9 | |
| Step1X-EditType=Open-source2025.05 | 14.9 | |
| Step1X-EditType=Open-source2026.03 | 14.9 |