List your company

Text-to-Image Generation on CC12M (val)

6.27FIDmmDiT-XL/2 (ours)
5.84048.740211.6414.5398Nov 5, 2024

Evaluation results

MethodLinks
mmDiT-XL/2 (ours)Resolution=5122024.116.27
mmDiT-XL/2 (ours)Resolution=2562024.116.79
mmDiT-XL/2 (SD3)Implementation=Our re-implementation, Resolution=2562024.117.54
UNet (SDXL)Implementation=Our re-implementation, Resolution=2562024.118.53
mmDiT-XL/2 (SD3)Implementation=Our re-implementation, Resolution=5122024.1111.24
UNet (SDXL)Implementation=Our re-implementation, Resolution=5122024.1112.56
UNet (SD/LDM-G4)Resolution=2562024.1117.01