UniPC: A Unified Predictor-Corrector Framework for Fast Sampling of Diffusion Models

TL;DR

UniPC is a unified predictor-corrector framework that accelerates diffusion model sampling, achieving high-quality images with fewer steps by improving accuracy without extra model evaluations.

Contribution

It introduces UniC and UniP, a unified corrector and predictor, forming a framework that enhances sampling speed and quality in diffusion models with arbitrary order.

Findings

Achieves 3.87 FID on CIFAR10 with 10 steps

Achieves 7.51 FID on ImageNet 256x256 with 10 steps

Significantly improves sampling quality in few-step scenarios

Abstract

Diffusion probabilistic models (DPMs) have demonstrated a very promising ability in high-resolution image synthesis. However, sampling from a pre-trained DPM is time-consuming due to the multiple evaluations of the denoising network, making it more and more important to accelerate the sampling of DPMs. Despite recent progress in designing fast samplers, existing methods still cannot generate satisfying images in many applications where fewer steps (e.g., $<$10) are favored. In this paper, we develop a unified corrector (UniC) that can be applied after any existing DPM sampler to increase the order of accuracy without extra model evaluations, and derive a unified predictor (UniP) that supports arbitrary order as a byproduct. Combining UniP and UniC, we propose a unified predictor-corrector framework called UniPC for the fast sampling of DPMs, which has a unified analytical form for any order and can significantly improve the sampling quality over previous methods, especially in extremely few steps. We evaluate our methods through extensive experiments including both unconditional and conditional sampling using pixel-space and latent-space DPMs. Our UniPC can achieve 3.87 FID on CIFAR10 (unconditional) and 7.51 FID on ImageNet 256$\times$256 (conditional) with only 10 function evaluations. Code is available at https://github.com/wl-zhao/UniPC.