Multi-image reasoning on OmniContext
8.96Single Scene Char ScoreGPT Image 1
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| GPT Image 1Setting=High2026.02 | 8.96 | 8.9 | 8.95 | 8.81 | 8.91 | 8.44 | 8.78 | |
| GPT-4o2025.12 | 8.9 | 9.07 | 8.95 | 8.54 | 8.44 | 8.6 | 8.75 | |
| Nano Banana2026.02 | 8.79 | 8.27 | 8.6 | 7.71 | 9.12 | 6.81 | 8.07 | |
| Loom2025.12 | 8.67 | 8.09 | 7.62 | 8.25 | 7.95 | 8.23 | 8.13 | |
| UniRef-Image-Edit2026.02 | 8.65 | 8.8 | 8.95 | 8.79 | 9.24 | 8.54 | 8.78 | |
| Echo-4o2025.12 | 8.62 | 8.07 | 7.5 | 8.29 | 8 | 8.08 | 8.09 | |
| Qwen-Image-Edit-2509Inference Framework=SELF2026.02 | 8.56 | 7.92 | 8.37 | 7.79 | 8.41 | 6.86 | 7.6 | |
| UniWorld-V22026.02 | 8.45 | 7.87 | 8.22 | 7.95 | 8.44 | 6.98 | 7.59 | |
| Echo-4o2026.02 | 8.34 | 8.13 | 8.14 | 8.11 | 8.27 | 7.77 | 7.95 | |
| Scone2026.02 | 8.34 | 8.24 | 8.14 | 8.3 | 8.52 | 7.63 | 8.01 | |
| OmniGen22026.02 | 8.17 | 7.26 | 7.03 | 7.56 | 7.63 | 6.64 | 7.28 | |
| FLUX.1 KontextVersion=dev2026.02 | 8.07 | — | — | — | 7.97 | — | — | |
| USO2026.02 | 8.03 | 3.32 | 6.1 | 4.56 | 7.55 | 5.09 | 5.35 | |
| UNO2026.02 | 7.15 | 3.56 | 6.46 | 4.9 | 6.72 | 4.76 | 5.14 | |
| BAGEL2026.02 | 7 | 5.32 | 6.69 | 6.74 | 7.04 | 5.73 | 6.03 | |
| OmniGen22025.12 | 6.38 | 7.11 | 7.13 | 7.45 | 6.71 | 7.04 | 6.97 | |
| Bagel2025.12 | 4.07 | 5.17 | 6.64 | 6.24 | 5.71 | 5.47 | 5.55 | |
| OmniGen2025.12 | 3.59 | 5.65 | 5.44 | 4.68 | 4.32 | 5.12 | 4.8 | |
| Gemini-2.0-flash2025.12 | 3.02 | 2.91 | 2.16 | 3.8 | 3.89 | 2.92 | 3.12 | |
| UNO2025.12 | 2.06 | 2.54 | 6.51 | 4.39 | 4.33 | 4.37 | 4.03 |