FADE: Fractional Anomalous Dynamics Extrapolation for Training-Free Diffusion Transformers Acceleration
Organizations: Sichuan University Chengdu, China · School of Artificial Intelligence Sichuan University Chengdu, China
Abstract
Diffusion Transformers (DiTs) have established themselves as the preeminent paradigm for high-fidelity generative modeling across various modalities, yet their practical deployment is hindered by the severe computational overhead of iterative sampling. While recent feature-caching methods mitigate this latency through feature reuse or forecasting based on integer-order operators, they remain limited in adapting to the non-stationary and highly curved dynamics of DiT latent trajectories. In this work, we reveal that DiT feature evolution exhibits anomalous diffusion characteristics and a certain degree of spatiotemporal heterogeneity, rendering static, integer-order forecasting operators inadequate and prone to error accumulation. To address this, we propose Fractional Anomalous Dynamics Extrapolation (FADE), a unified framework for training-free diffusion transformers acceleration. FADE models DiT feature evolution via anomalous diffusion and employs a Mittag--Leffler fractional operator to better adapt to high-curvature and non-linear latent trajectories of DiTs. Leveraging the empirical observation that these complex dynamics maintain cross-sample stability, we further construct a Look-Up Table with multiple dimensions through offline profiling. This enables negligible LUT-retrieval overhead, spatiotemporally adaptive parameter retrieval during online inference. Extensive evaluations across diverse architectures and modalities demonstrate that FADE achieves state-of-the-art acceleration, preserving exceptional generation fidelity under aggressive skip intervals with negligible latency overhead. Our project page is available at https://youngkinlon.github.io/fade-project-page/.
Figures & tables
| Method | Lat. (s) | FLOPs (T) | Spd. | FID | sFID | IS |
|---|---|---|---|---|---|---|
| DDIM-50 steps Song et al. (2022) | 1.18 | 23.74 | 1.00 | 2.29 | 4.29 | 237.31 |
| DDIM-25 steps Song et al. (2022) | 0.62 | 11.87 | 2.00 | 2.91 | 4.57 | 232.22 |
| DDIM-15 steps Song et al. (2022) | 0.42 | 7.12 | 3.33 | 5.09 | 6.11 | 205.54 |
| FORA( =4) Selvaraju et al. (2024) | 0.45 | 6.66 | 3.56 | 4.73 | 8.45 | 214.20 |
| Taylor ( =4, =4) Liu et al. (2025b) | 0.61 | 6.66 | 3.56 | 2.59 | 5.18 | 232.90 |
| HiCache ( =4, =4) Feng et al. (2026) | 0.57 | 6.66 | 3.56 | 2.51 | 5.20 | 232.72 |
| Method | Latency(s) | FLOPs(T) | Speed | Img. Reward | CLIP Score | Cycle Reward | PSNR | SSIM | LPIPS |
|---|---|---|---|---|---|---|---|---|---|
| FLUX.1 [dev] - 50 steps † | 17.18 | 3719.50 | 1.00 | 0.9898 | 27.4611 | 0.8800 | 1.0000 | 0.0000 | |
| FLUX.1 [dev] - 25 steps | 8.80 | 1859.75 | 2.00 | 0.9375 | 27.3388 | 0.8466 | 19.2599 | 0.7749 | 0.3583 |
| FLUX.1 [dev] - 20 steps | 7.12 | 1487.80 | 2.62 | 0.9387 | 27.2505 | 0.8466 | 18.1371 | 0.7484 | 0.4006 |
| FORA ( =5) Selvaraju et al. (2024) | 5.01 | 893.54 | 4.16 | 0.8253 | 27.2377 | 0.8353 | 15.9899 | 0.6677 | 0.5206 |
| TaylorSeer ( =5, m=1) Liu et al. (2025b) | 5.14 | 893.54 | 4.16 | 0.9919 | 27.5143 | 0.8540 | 18.4497 | 0.7424 | 0.4146 |
| HiCache ( =5, m=1) Feng et al. (2026) | 5.12 | 893.54 | 4.16 | 0.9350 | 27.4034 | 0.8346 | 19.2595 | 0.7503 | 0.3911 |
| Method | Lat. (s) | FLOPs (T) | Spd. | VBench (%) |
|---|---|---|---|---|
| Original (50 steps) | 190.2 | 29773 | 1.00 | 80.68 |
| DDIM (22% steps) Song et al. (2022) | 42.7 | 6550 | 4.55 | 78.73 |
| FORA (N=6) Selvaraju et al. (2024) | 37.1 | 5359 | 5.56 | 78.86 |
| TaylorSeer (N=6, m=1) Liu et al. (2025b) | 37.8 | 5359 | 5.56 | 79.87 |
| HiCache (N=6, m=1) Feng et al. (2026) | 37.8 | 5359 | 5.56 | 79.93 |
| FADE (N=6, m=1) | 37.9 | 5359 | 5.56 | 80.12 |
Appendix figures & tables7 assets
Supplementary material from the paper’s appendix.
Appendix
| Network Phase | Layer Indices | Attention Module (Attn) | ||
|---|---|---|---|---|
| Mean | Std ( ) | MAD | ||
| Early Layers | Layers 0–9 | 0.64 | 0.144 | 0.128 |
| Middle Layers | Layers 10–18 | 0.96 | 0.129 | 0.119 |
| Deep Layers | Layers 19–27 | 0.46 | 0.128 | 0.109 |
| Global Summary | All Layers | 0.68 | 0.120 | 0.123 |
| Method | Speed | IR | CLIP |
|---|---|---|---|
| FADE (fixed ) | 4.99 | 0.9883 | 27.5061 |
| FADE (fixed ) | 4.99 | 1.0016 | 27.6207 |
| FADE (fixed ) | 4.99 | 0.9978 | 27.6239 |
| FADE (fixed ) | 4.99 | 0.9936 | 27.6617 |
| FADE (dynamic fit ) | 4.99 | 0.9948 | 27.7077 |
| Method | FID | sFID |
|---|---|---|
| DDIM-50 | 2.29 | 4.29 |
| HiCache | 2.51 | 5.20 |
| Taylor | 2.590 | 5.18 |
| FADE (fixed ) | 2.49 | 4.81 |
| FADE (Bernoulli-fit ) | 2.51 | 5.14 |
| Model | Cache interval | Refinement factor |
|---|---|---|
| DiT-XL/2 | 4 | 0.6 |
| DiT-XL/2 | 5 | 0.6 |
| DiT-XL/2 | 6 | 0.7 |
| FLUX.1-dev | 5 | 1.3 |
| FLUX.1-dev | 6 | 1.0 |
| FLUX.1-dev | 7 | 1.1 |
Diffusion Model Acceleration
Diffusion model acceleration reduces the training and sampling cost of iterative generative models while preserving output quality.
See all 152 papers