PixArt-Σ

Reported on 4 benchmarks across 4 tasks · 1 paper · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio2 results

10-shot image generationonGenEval
Overall· 2024-03-07
0.53
best: 0.95 (SD3.5-Medium+Flow-GRPO)
SOTA
PixArt-Σ: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation arXiv:2403.04692
1 Image, 2*2 StitchionGenEval
Overall· 2024-03-07
0.53
best: 0.95 (SD3.5-Medium+Flow-GRPO)
SOTA
PixArt-Σ: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation arXiv:2403.04692

Medical1 result

Image GenerationonGenEval
Overall· 2024-03-07
0.53
best: 0.95 (SD3.5-Medium+Flow-GRPO)
SOTA
PixArt-Σ: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation arXiv:2403.04692

Natural Language Processing1 result

Text-to-Image GenerationonGenEval
Overall· 2024-03-07
0.53
best: 0.95 (SD3.5-Medium+Flow-GRPO)
SOTA
PixArt-Σ: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation arXiv:2403.04692