mlsanityData to 14 Oct 2026

OMS-DPM: Optimizing the Model Schedule for Diffusion Probabilistic Models

arXiv 2306.08860 ↗First posted 15 Jun 2023cs.LGcited by 45

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Diffusion probabilistic models (DPMs) are a new class of generative models that have achieved state-of-the-art generation quality in various domains. Despite the promise, one major drawback of DPMs is the slow generation speed due to the large number of neural network evaluations required in the generation process. In this paper, we reveal an overlooked dimension -- model schedule -- for optimizing the trade-off between generation quality and speed. More specifically, we observe that small models, though having worse generation quality when used alone, could outperform large models in certain generation steps. Therefore, unlike the traditional way of using a single model, using different models in different generation steps in a carefully designed model schedule could potentially improve generation quality and speed simultaneously. We design OMS-DPM, a predictor-based search algorithm, to optimize the model schedule given an arbitrary generation time budget and a set of pre-trained models. We demonstrate that OMS-DPM can find model schedules that improve generation quality and speed than prior state-of-the-art methods across CIFAR-10, CelebA, ImageNet, and LSUN datasets. When applied to the public checkpoints of the Stable Diffusion model, we are able to accelerate the sampling by 2× while maintaining the generation quality.

Key facts. Cited by 45 papers in the map (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 35 of its 42 references that are in the map. First posted on arXiv 15 Jun 2023.

Builds on

35 of 42 references.

Closest ancestors · 17

Same direction, newest first

  1. 2022Diffusion Probabilistic Model Made Slim27 Nov
  2. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps2 Jun
  3. 2022Subspace Diffusion Generative Models3 May
  4. 2022Fast Sampling of Diffusion Models with Exponential Integrator29 Apr
  5. 2022Perception Prioritized Training of Diffusion Models1 Apr
  6. 2022Pseudo Numerical Methods for Diffusion Models on Manifolds20 Feb
  7. 2022Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Quality11 Feb
  8. 2022Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models17 Jan
  9. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  10. 2021Soft Truncation: A Universal Training Technique of Score-based Diffusion Model for High Precision Score Estimation10 Jun
  11. 2021Learning to Efficiently Sample from Diffusion Probabilistic Models7 Jun
  12. 2021Diffusion Models Beat GANs on Image Synthesis11 May
5 more
  1. 2021Improved Denoising Diffusion Probabilistic Models18 Feb
  2. 2021Maximum Likelihood Training of Score-Based Diffusion Models22 Jan
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  4. 2020Denoising Diffusion Implicit Models6 Oct
  5. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 18

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  4. 2022Generative Adversarial Networkscited by 5,627
  5. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  6. 2022Video Diffusion Modelscited by 2,305
  7. 2021Image Super-Resolution via Iterative Refinementcited by 1,511
  8. 2016Neural Architecture Search with Reinforcement Learningcited by 1,499
  9. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,354
  10. 2022eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoiserscited by 931
  11. 2018Regularized Evolution for Image Classifier Architecture Searchcited by 909
  12. 2021Diffusion Probabilistic Models for 3D Point Cloud Generationcited by 735
6 more
  1. 2020WaveGrad: Estimating Gradients for Waveform Generationcited by 622
  2. 2021SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Modelscited by 415
  3. 2017Population Based Training of Neural Networkscited by 367
  4. 2018Neural Architecture Optimizationcited by 167
  5. 2020A Generic Graph-based Neural Architecture Encoding Scheme for Predictor-based NAScited by 55
  6. 2018OBOE: Collaborative Filtering for AutoML Model Selectioncited by 31

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 7; 2024: 20; 2025: 10; 2026: 8.

45 papers.

Sort:

By area:

  1. 2023DeepCache: Accelerating Diffusion Models for Freecited by 359
  2. 2024Delta-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformerscited by 175
  3. 2024Learning-to-Cache: Accelerating Diffusion Transformer via Layer Cachingcited by 142
  4. 2024ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generationcited by 115
  5. 2024DiTFastAttn: Attention Compression for Diffusion Transformer Modelscited by 115
  6. 2024Accelerating Diffusion Sampling with Optimized Time Stepscited by 67
  7. 2024Efficient Expert Pruning for Sparse Mixture-of-Experts Language Models: Enhancing Performance and Reducing Inference Costscited by 53
  8. 2024Attention-Driven Training-Free Efficiency Enhancement of Diffusion Modelscited by 38
  9. 2023Faster Diffusion: Rethinking the Role of the Encoder for Diffusion Model Inferencecited by 34
  10. 2024Denoising Task Difficulty-based Curriculum for Training Diffusion Modelscited by 27
  11. 2023Denoising Diffusion Step-aware Modelscited by 24
  12. 2026Diversity-Preserved Distribution Matching Distillation for Fast Visual Synthesiscited by 20
33 more
  1. 2024HarmoniCa: Harmonizing Training and Inference for Better Feature Caching in Diffusion Transformer Accelerationcited by 17
  2. 2023Towards More Accurate Diffusion Model Acceleration with A Timestep Tunercited by 14
  3. 2023A Unified Sampling Framework for Solver Searching of Diffusion Probabilistic Modelscited by 14
  4. 2024Not All Prompts Are Made Equal: Prompt-based Pruning of Text-to-Image Diffusion Modelscited by 13
  5. 2025CacheQuant: Comprehensively Accelerated Diffusion Modelscited by 13
  6. 2024Mixture of Efficient Diffusion Experts Through Automatic Interval and Sub-Network Selectioncited by 12
  7. 2024FlashEval: Towards Fast and Accurate Evaluation of Text-to-image Diffusion Generative Modelscited by 11
  8. 2024SparseDiT: Token Sparsification for Efficient Diffusion Transformercited by 10
  9. 2025LeMiCa: Lexicographic Minimax Path Caching for Efficient Diffusion-Based Video Generationcited by 10
  10. 2026MeanCache: From Instantaneous to Average Velocity for Accelerating Flow Matching Inferencecited by 10
  11. 2024ENAT: Rethinking Spatial-temporal Interactions in Token-based Image Synthesiscited by 8
  12. 2024Decouple-Then-Merge: Finetune Diffusion Models as Multi-Task Learningcited by 7
  13. 2024Linear Combination of Saved Checkpoints Makes Consistency and Diffusion Models Bettercited by 6
  14. 2023Your Student is Better Than Expected: Adaptive Teacher-Student Collaboration for Text-Conditional Diffusion Modelscited by 5
  15. 2024Remix-DiT: Mixing Diffusion Transformers for Multi-Expert Denoisingcited by 5
  16. 2025Compact Attention: Exploiting Structured Spatio-Temporal Sparsity for Fast Video Generationcited by 5
  17. 2025Taming Flow-based I2V Models for Creative Video Editingcited by 5
  18. 2024Diffusion Model Compression for Image-to-Image Translationcited by 4
  19. 2026Semantic Routing: Exploring Multi-Layer LLM Feature Weighting for Diffusion Transformerscited by 4
  20. 2023Adaptive Training Meets Progressive Scaling: Elevating Efficiency in Diffusion Modelscited by 3
  21. 2024Flexiffusion: Segment-wise Neural Architecture Search for Flexible Denoising Schedulecited by 1
  22. 2025Efficient Personalization of Quantized Diffusion Model without Backpropagationcited by 1
  23. 2025Flexiffusion: Training-Free Segment-Wise Neural Architecture Search for Efficient Diffusion Modelscited by 1
  24. 2025BADiff: Bandwidth Adaptive Diffusion Modelcited by 1
  25. 2026AdaGen: Learning Adaptive Policy for Image Synthesiscited by 1
  26. 2026Not All Denoising Steps Are Equal: Model Scheduling for Faster Masked Diffusion Language Modelscited by 1
  27. 2026From Atoms to Entropy: Optimal Noise Allocation for Diffusion Training in the Convex Regimecited by 1
  28. 2024FRRffusion: Unveiling Authenticity with Diffusion-Based Face Retouching Reversalcited by 0
  29. 20252ndMatch: Finetuning Pruned Diffusion Models via Second-Order Jacobian Matchingcited by 0
  30. 2025MARch'e: Fast Masked Autoregressive Image Generation with Cache-Aware Attentioncited by 0
  31. 2025FlowBlending: Stage-Aware Multi-Model Sampling for Fast and High-Fidelity Video Generationcited by 0
  32. 2026TMPDiff: Temporal Mixed-Precision for Diffusion Modelscited by 0
  33. 2026InstantFusion: A Shared Latent Interface for Heterogeneous Image Generatorscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024An overview of diffusion models for generative artificial intelligence
  2. 2023Fast Diffusion Probabilistic Model Sampling through the lens of Backward Error Analysis
  3. 2024Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
  4. 2022Estimating the Optimal Covariance with Imperfect Mean in Diffusion Probabilistic Models
  5. 2024Denoising Diffusion Probabilistic Models in Six Simple Steps
  6. 2025Hierarchical Schedule Optimization for Fast and Robust Diffusion Model Sampling
  7. 2023The Missing U for Efficient Diffusion Models
  8. 2024Learning to Discretize Denoising Diffusion ODEs