Data Party THU

Aug 9, 2025 · Artificial Intelligence

How SADA Boosts Diffusion Model Sampling Speed by Up to 1.8× Without Losing Quality

The paper introduces SADA (Stability‑guided Adaptive Diffusion Acceleration), a novel paradigm that dynamically allocates sparsity per token using a unified stability criterion, enabling efficient ODE‑based sampling for diffusion and flow‑matching models, achieving up to 1.8× speedup with negligible fidelity loss across SD‑2, SDXL, Flux, ControlNet and MusicLDM.