mlsanityData to 7 Oct 2026

FreqCa: Accelerating Diffusion Models via Frequency-Aware Caching

arXiv 2510.08669 ↗First posted 9 Oct 2025cs.LG cs.AI cs.CVcited by 16

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

The application of diffusion transformers is suffering from their significant inference costs. Recently, feature caching has been proposed to solve this problem by reusing features from previous timesteps, thereby skipping computation in future timesteps. However, previous feature caching assumes that features in adjacent timesteps are similar or continuous, which does not always hold in all settings. To investigate this, this paper begins with an analysis from the frequency domain, which reveal that different frequency bands in the features of diffusion models exhibit different dynamics across timesteps. Concretely, low-frequency components, which decide the structure of images, exhibit higher similarity but poor continuity. In contrast, the high-frequency bands, which decode the details of images, show significant continuity but poor similarity. These interesting observations motivate us to propose Frequency-aware Caching (FreqCa) which directly reuses features of low-frequency components based on their similarity, while using a second-order Hermite interpolator to predict the volatile high-frequency ones based on its continuity. Besides, we further propose to cache Cumulative Residual Feature (CRF) instead of the features in all the layers, which reduces the memory footprint of feature caching by 99%. Extensive experiments on FLUX.1-dev, FLUX.1-Kontext-dev, Qwen-Image, and Qwen-Image-Edit demonstrate its effectiveness in both generation and editing. Codes are available in the supplementary materials and will be released on GitHub.

Key facts. Cited by 16 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 43 of its 51 references that are in the map. First posted on arXiv 9 Oct 2025.

Builds on

43 of 51 references.

Closest ancestors · 29

Same direction, newest first

  1. 2025SpeCa: Accelerating Diffusion Transformers with Speculative Feature Caching15 Sep
  2. 2025HiCache: A Plug-in Scaled-Hermite Upgrade for Taylor-Style Cache-then-Forecast Diffusion Acceleration23 Aug
  3. 2025OmniCache: A Trajectory-Oriented Global Perspective on Training-Free Cache Reuse for Diffusion Transformer Models22 Aug
  4. 2025Forecast then Calibrate: Feature Caching as ODE for Efficient Diffusion Transformers22 Aug
  5. 2025Accelerating Diffusion Transformer via Increment-Calibrated Caching with Channel-Aware Singular Value Decomposition9 May
  6. 2025From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers10 Mar
  7. 2025Region-Adaptive Sampling for Diffusion Transformers14 Feb
  8. 2025Accelerating Diffusion Transformer via Error-Optimized Cache31 Jan
  9. 2025Ditto: Accelerating Diffusion Model via Temporal Value Similarity20 Jan
  10. 2024Token Pruning for Caching Better: 9 Times Acceleration on Stable Diffusion for Free31 Dec
  11. 2024Rethinking Token-wise Feature Caching: Accelerating Diffusion Transformers with Dual Feature Caching25 Dec
  12. 2024DiP-GO: A Diffusion Pruner via Few-step Gradient Optimization22 Oct
17 more
  1. 2024Accelerating Diffusion Transformers with Token-wise Feature Caching5 Oct
  2. 2024FORA: Fast-Forward Caching in Diffusion Transformer Acceleration1 Jul
  3. 2024Delta-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers3 Jun
  4. 2023DeepCache: Accelerating Diffusion Models for Free1 Dec
  5. 2023DPM-Solver-v3: Improved Diffusion ODE Solver with Empirical Model Statistics20 Oct
  6. 2023Structural Pruning for Diffusion Models18 May
  7. 2023Token Merging for Fast Stable Diffusion30 Mar
  8. 2023Consistency Models2 Mar
  9. 2023Q-Diffusion: Quantizing Diffusion Models8 Feb
  10. 2022Scalable Diffusion Models with Transformers19 Dec
  11. 2022Post-training Quantization on Diffusion Models28 Nov
  12. 2022DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models2 Nov
  13. 2022On Distillation of Guided Diffusion Models6 Oct
  14. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps2 Jun
  15. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  16. 2020Denoising Diffusion Implicit Models6 Oct
  17. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 14

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  5. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  6. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,526
  7. 2021CLIPScore: A Reference-free Evaluation Metric for Image Captioningcited by 2,388
  8. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,198
  9. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Spacecited by 960
  10. 2024Open-Sora: Democratizing Efficient Video Production for Allcited by 637
  11. 2023SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Secondscited by 227
  12. 2024Real-Time Video Generation with Pyramid Attention Broadcastcited by 140
2 more
  1. 2023Token Fusion: Bridging the Gap between Token Pruning and Token Mergingcited by 123
  2. 2024FasterCache: Training-Free Video Diffusion Model Acceleration with High Qualitycited by 115

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 15.

16 papers.

Sort:

By area:

  1. 2025A Survey on Cache Methods in Diffusion Models: Toward Efficient Multi-Modal Generationcited by 14
  2. 2026SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Modelscited by 10
  3. 2026Visual Generation in the New Era: An Evolution from Atomic Mapping to Agentic World Modelingcited by 5
  4. 2026LESA: Learnable Stage-Aware Predictors for Diffusion Model Accelerationcited by 4
  5. 2026TAP: A Token-Adaptive Predictor Framework for Training-Free Diffusion Accelerationcited by 3
  6. 2026Spectral Progressive Diffusion for Efficient Image and Video Generationcited by 3
  7. 2026ResilPhase: Plug-and-Play Phase Mapping and Noise-Resilient Macro-Trajectory Extrapolation for Diffusion Accelerationcited by 3
  8. 2026Beyond Fixed Formulas: Data-Driven Linear Predictor for Efficient Diffusion Modelscited by 2
  9. 2026Accelerating Frequency Domain Diffusion Models with Error-Feedback Event-Driven Cachingcited by 1
  10. 2026SpecEdit: Training-Free Acceleration for Diffusion based Image Editing via Semantic Lockingcited by 1
  11. 2026HEART: Exploiting Head Heterogeneity in Sparse Attention for Video Diffusioncited by 1
  12. 2026BeCARE: Budgeted Cache Refresh for Diffusion Transformer Accelerationcited by 1
4 more
  1. 2026OTCache: Optimal Transport for Geometry-Aware Caching in Diffusion Modelscited by 1
  2. 2026SyncCache: Exploiting Asymmetric Dynamics for Fast Audio-Driven Portrait Animationcited by 0
  3. 2026LinCa: Accelerating Diffusion Models via Learnable Decomposed Feature Cachingcited by 0
  4. 2026AViTS: Adaptive Spatiotemporal Token Selection for Efficient Dynamic-Resolution Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025AB-Cache: Training-Free Acceleration of Diffusion Models via Adams-Bashforth Cached Feature Reuse
  2. 2026Forecast the Principal, Stabilize the Residual: Subspace-Aware Feature Caching for Efficient Diffusion Transformers
  3. 2025Compute Only 16 Tokens in One Timestep: Accelerating Diffusion Transformers with Cluster-Driven Feature Caching
  4. 2026Frequency-Aware Error-Bounded Caching for Accelerating Diffusion Transformers
  5. 2025ProCache: Constraint-Aware Feature Caching with Selective Computation for Diffusion Transformer Acceleration
  6. 2026OmniCache: Multidimensional Hierarchical Feature Caching For Diffusion Models
  7. 2025Let Features Decide Their Own Solvers: Hybrid Feature Caching for Diffusion Transformers
  8. 2025H2-Cache: A Novel Hierarchical Dual-Stage Cache for High-Performance Acceleration of Generative Diffusion Models