Text-to-image generation on CVTG
84AccuracyThinkGen*
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ThinkGen*Model Category=Und. and Gen., CoT reasoning=true2025.12 | 84 | 0.94 | |
| ThinkGenModel Category=Und. and Gen., CoT reasoning=false2025.12 | 80 | 0.91 | |
| TextCrafterModel Category=Gen. Only2025.12 | 76 | 0.9 | |
| SD3-MediumModel Category=Gen. Only2025.12 | 65 | 0.84 | |
| OmniGen2Model Category=Und. and Gen.2025.12 | 52 | 0.77 | |
| FLUX.1-devModel Category=Gen. Only2025.12 | 49 | 0.68 | |
| BAGELModel Category=Und. and Gen.2025.12 | 35 | 0.65 | |
| BAGEL*Model Category=Und. and Gen.2025.12 | 11 | 0.39 |