Forecast the Principal, Stabilize the Residual: Subspace-Aware Feature Caching for Efficient Diffusion Transformers

TL;DR

This paper introduces SVD-Cache, a subspace-aware feature caching method for diffusion transformers that significantly accelerates inference by decomposing features into principal and residual subspaces, enabling near-lossless speedup.

Contribution

The paper reveals the distinct behaviors of principal and residual subspaces in diffusion features and proposes SVD-Cache, a novel caching framework leveraging SVD and EMA prediction for efficient inference.

Findings

Achieves 5.55× speedup on FLUX and HunyuanVideo models.

Maintains near-lossless quality across diverse models and methods.

Compatible with model acceleration techniques like distillation and quantization.

Abstract

Diffusion Transformer (DiT) models have achieved unprecedented quality in image and video generation, yet their iterative sampling process remains computationally prohibitive. To accelerate inference, feature caching methods have emerged by reusing intermediate representations across timesteps. However, existing caching approaches treat all feature components uniformly. We reveal that DiT feature spaces contain distinct principal and residual subspaces with divergent temporal behavior: the principal subspace evolves smoothly and predictably, while the residual subspace exhibits volatile, low-energy oscillations that resist accurate prediction. Building on this insight, we propose SVD-Cache, a subspace-aware caching framework that decomposes diffusion features via Singular Value Decomposition (SVD), applies exponential moving average (EMA) prediction to the dominant low-rank components, and directly reuses the residual subspace. Extensive experiments demonstrate that SVD-Cache achieves near-lossless across diverse models and methods, including 5.55$\times$ speedup on FLUX and HunyuanVideo, and compatibility with model acceleration techniques including distillation, quantization and sparse attention. Our code is in supplementary material and will be released on Github.