Phased Consistency Models

TL;DR

Phased Consistency Models (PCMs) improve high-resolution, text-conditioned image and video generation by addressing limitations of Latent Consistency Models, achieving superior multi-step refinement and competitive one-step results.

Contribution

Introduction of Phased Consistency Models that generalize and enhance Latent Consistency Models for better high-resolution, text-conditioned image and video generation.

Findings

PCMs outperform LCMs in 1--16 step generation.

PCMs achieve comparable 1-step results to state-of-the-art methods.

PCMs enable state-of-the-art few-step text-to-video generation.

Abstract

Consistency Models (CMs) have made significant progress in accelerating the generation of diffusion models. However, their application to high-resolution, text-conditioned image generation in the latent space remains unsatisfactory. In this paper, we identify three key flaws in the current design of Latent Consistency Models (LCMs). We investigate the reasons behind these limitations and propose Phased Consistency Models (PCMs), which generalize the design space and address the identified limitations. Our evaluations demonstrate that PCMs outperform LCMs across 1--16 step generation settings. While PCMs are specifically designed for multi-step refinement, they achieve comparable 1-step generation results to previously state-of-the-art specifically designed 1-step methods. Furthermore, we show the methodology of PCMs is versatile and applicable to video generation, enabling us to train the state-of-the-art few-step text-to-video generator. Our code is available at https://github.com/G-U-N/Phased-Consistency-Model.