Text-to-Image Generation on GenExam Relaxed Score
87.8Math Score (GenExam Relaxed)Nano Banana 2
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Nano Banana 2Model Type=Closed-Source2026.07 | 87.8 | 95.7 | 90 | 95.2 | 94.8 | 88.8 | 95.8 | 94.2 | 86.9 | 97.3 | 92.6 | |
| Nano Banana ProModel Type=Closed-Source2026.07 | 86.3 | 95.1 | 88.7 | 95.9 | 96.5 | 91.7 | 95.1 | 97.2 | 91 | 99.9 | 93.7 | |
| GPT-Image-2Model Type=Closed-Source2026.07 | 85.2 | 95.6 | 92 | 97.5 | 97.6 | 93.3 | 96.5 | 97.7 | 89.1 | 97.1 | 93.8 | |
| Seedream 5.0Model Type=Closed-Source2026.07 | 82.9 | 85.7 | 81.1 | 89.9 | 89.5 | 85.2 | 91.2 | 94.7 | 76.7 | 87 | 86.4 | |
| GPT-Image-1.5Model Type=Closed-Source2026.07 | 65.8 | 85.4 | 78.1 | 91.9 | 92.5 | 75.8 | 86.4 | 85.5 | 70.8 | 90.9 | 82.3 | |
| GPT-Image-1Model Type=Closed-Source2026.07 | 52 | 66.4 | 53.4 | 74.6 | 73.9 | 55.6 | 65.5 | 65.8 | 52.6 | 67.4 | 62.6 | |
| FLUX.2 MaxModel Type=Closed-Source2026.07 | 49.1 | 63.2 | 54 | 74.5 | 76.3 | 56.5 | 68.9 | 61.5 | 47 | 68 | 61.9 | |
| Seedream 4.5Model Type=Closed-Source2026.07 | 44.7 | 63.4 | 48.9 | 75.8 | 67.6 | 57.9 | 69.7 | 67.3 | 38 | 55 | 58.8 | |
| discipline-informed reasoning-generation modelModel Type=Open-Source2026.07 | 39.3 | 51.2 | 53.7 | 61.9 | 63.1 | 41.8 | 58.4 | 60.5 | 44 | 60.9 | 51.4 | |
| FLUX.2 devModel Type=Open-Source2026.07 | 31.6 | 42.7 | 33.2 | 54.8 | 62.6 | 31.1 | 48.9 | 43.6 | 33.4 | 47.5 | 42.9 | |
| Qwen-Image-2512Model Type=Open-Source2026.07 | 27.9 | 41.3 | 23.2 | 44.4 | 56.6 | 24.1 | 42.9 | 32.3 | 28.3 | 37 | 35.8 | |
| BAGEL (thinking)Model Type=Open-Source2026.07 | 11.7 | 13.8 | 11.9 | 15.2 | 28.5 | 6.2 | 10.7 | 6.3 | 14.7 | 16 | 13.5 | |
| Show-o2-7BModel Type=Open-Source2026.07 | 10.8 | 11.9 | 4.8 | 12.8 | 33.3 | 4.7 | 11.8 | 7 | 8.8 | 14.5 | 12 |