TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows
Open paper
Benchmark
Model
Rank
Results
text-to-image-generation-on-dpg-bench
Qwen-Image-TwinFlow (full, longer training)
#22
DPG-Bench Score: 87.8
text-to-image-generation-on-geneval
Qwen-Image-TwinFlow (full, longer training)
#22
Overall: 0.9
Rank counts only results with a code link.