Text-to-Image Synthesis on Aesthetic-Eval 2K resolution (test)
38.57gFIDFlux-SGA-SAM2
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Flux-SGA-SAM2Vision Foundation Model=SAM22026.05 | 38.57 | 34.46 | 6.38 | 0.79 | 9.98 | |
| Flux-SGA-DINOv3Vision Foundation Model=DINOv32026.05 | 39.39 | 34.53 | 6.39 | 0.79 | 10.49 | |
| Flux-WLFBaseline=Diffusion-4K2026.05 | 39.49 | 34.41 | 6.37 | 0.61 | 13.6 | |
| Flux-SGA-DINOv2Vision Foundation Model=DINOv22026.05 | 39.75 | 34.47 | 6.38 | 0.81 | 10.06 | |
| Flux2026.05 | 50.57 | 30.41 | 6.36 | 0.58 | 14.8 |