cs.CVNov 18, 2026

CalibCache: Cache-to-Full Feature Dynamics Calibration for Diffusion Transformers

Authors: Mingyu Liang, Dingkun Xu, Jingwei Xu

Organizations: School of Computer Science, Nanjing University

Abstract

Diffusion Transformers generate high-quality images, but iterative sampling remains computationally expensive. Feature caching reduces the number of full network evaluations, yet its approximation error can become more severe under short sampling schedules, where fewer full-compute steps and a coarser sampling grid can make history-based extrapolation less reliable. In this paper, we propose CalibCache, a lightweight offline calibration method that corrects systematic discrepancies between cached and full-compute feature dynamics. Specifically, CalibCache calibrates feature acceleration, defined as the second-order finite difference across sampling steps. Using only a single pair of full-compute and naive-cache trajectories at each setting, CalibCache fits scalar acceleration maps offline by centered least squares. These maps are then applied during sampling without additional full-network evaluations or online optimization. Across FLUX.1-dev and Qwen-Image on DrawBench and DiT-XL/2 on ImageNet-256, CalibCache consistently improves fidelity or generation quality over TaylorSeer under 20-step sampling at cache intervals N=3,4,5N=3,4,5. At N=5N=5, it reduces FLUX.1-dev LPIPS from 0.5630.563 to 0.4290.429 while achieving a 2.86×2.86\times denoising speedup over full-compute sampling, and lowers DiT-XL/2 FID from 16.3916.39 to 10.3710.37 relative to TaylorSeer.

Figures & tables

Appendix figures & tables18 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Trajectory-Consistent Calibration for Cache-Accelerated Diffusion Models

    May 24, 2026Mingyu Liang, Dingkun Xu, Jingwei XuDiffusion TransformerDiffusion-Based Image Generation

  2. LearniBridge: Learnable Calibration of Feature Caching for Diffusion Models Acceleration

    Jun 25, 2026Xuyue Huang, Zhe Chen, Wang Shen +1Video Diffusion ModelsDiffusion Transformer

  3. Rethinking What to Cache in Few-Step Diffusion Transformers: Solver-Aware Target Selection

    Oct 2, 2026Shuo Yang, Lihao Fang, Yi Zhang +5Diffusion Model CachingDiffusion Model Inference Acceleration