Text-to-Image Generation on UniGenBench++ (Short and Long Metrics)
68.85Score (Short)DALLE-3
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DALLE-32026.05 | 68.85 | 70.82 | |
| ARR-RPOVariant=w/ RPO-Gemini 3.1 Pro-ARR2026.05 | 65.89 | 72.93 | |
| ARR-RPOVariant=w/ RPO-GPT-5-ARR2026.05 | 65.36 | 72.41 | |
| ARR-RPOVariant=w/ RPO-Gemini 3.1 Pro2026.05 | 64.76 | 72.13 | |
| ARR-RPOVariant=w/ RPO-GPT-52026.05 | 64.22 | 71.78 | |
| ARR-RPOVariant=w/ RPO-Qwen3vl-8B-ARR2026.05 | 64.17 | 71.82 | |
| ARR-RPOVariant=w/ RPO-Qwen3vl-8B2026.05 | 63.28 | 71.05 | |
| OmniGen22026.05 | 63.09 | 71.39 | |
| Show-o22026.05 | 61.9 | 70.33 | |
| Janus-Pro-7B2026.05 | 61.36 | 71.11 | |
| FLUX.1-Dev2026.05 | 60.97 | 69.42 | |
| BAGEL2026.05 | 59.91 | 71.26 | |
| BLIP3o-4B2026.05 | 59.57 | 61.01 | |
| JanusFlow2026.05 | 47.1 | 54.8 | |
| Emu32026.05 | 45.42 | 50.59 | |
| SDXL2026.05 | 40.22 | 41.48 |