LinCa: Accelerating Diffusion Models via Learnable Decomposed Feature Caching
Open paper
Benchmark
Model
Rank
Results
text-to-video-generation-on-vbench
LinCa (Wan2.1-1.3B, N=6)
#31
VBench Total Score: 82.56
text-to-image-generation-on-geneval
LinCa (Qwen-Image-Lightning, N=3)
#68
Overall: 0.85
Rank counts only results with a code link.