# LD-Pruner: Efficient Pruning of Latent Diffusion Models using Task-Agnostic Insights

arXiv: https://arxiv.org/abs/2404.11936 · HTML: https://mlsanity.com/paper/2404.11936/

## Key facts

- Cited by 65 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 26 of its 37 references that are in the map.
- First posted on arXiv 18 Apr 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 26 of 26)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Pruning Filters for Efficient ConvNets](https://mlsanity.com/paper/1608.08710/) (2016)
- [CogView: Mastering Text-to-Image Generation via Transformers](https://mlsanity.com/paper/2105.13290/) (2021)
- [Learning Efficient Convolutional Networks through Network Slimming](https://mlsanity.com/paper/1708.06519/) (2017)
- [AudioLDM: Text-to-Audio Generation with Latent Diffusion Models](https://mlsanity.com/paper/2301.12503/) (2023)
- [LION: Latent Point Diffusion Models for 3D Shape Generation](https://mlsanity.com/paper/2210.06978/) (2022)
- [Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors](https://mlsanity.com/paper/2203.13131/) (2022)
- [SVDiff: Compact Parameter Space for Diffusion Fine-Tuning](https://mlsanity.com/paper/2303.11305/) (2023)
- [CogView2: Faster and Better Text-to-Image Generation via Hierarchical Transformers](https://mlsanity.com/paper/2204.14217/) (2022)
- [Q-Diffusion: Quantizing Diffusion Models](https://mlsanity.com/paper/2302.04304/) (2023)
- [SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds](https://mlsanity.com/paper/2306.00980/) (2023)
- [Fr'echet Audio Distance: A Metric for Evaluating Music Enhancement Algorithms](https://mlsanity.com/paper/1812.08466/) (2018)
- [Token Merging for Fast Stable Diffusion](https://mlsanity.com/paper/2303.17604/) (2023)
- [DepGraph: Towards Any Structural Pruning](https://mlsanity.com/paper/2301.12900/) (2023)
- [GALIP: Generative Adversarial CLIPs for Text-to-Image Synthesis](https://mlsanity.com/paper/2301.12959/) (2023)
- [HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmark](https://mlsanity.com/paper/2103.10584/) (2021)
- [BK-SDM: A Lightweight, Fast, and Cheap Version of Stable Diffusion](https://mlsanity.com/paper/2305.15798/) (2023)
- [The Framework Tax: Disparities Between Inference Efficiency in NLP Research and Deployment](https://mlsanity.com/paper/2302.06117/) (2023)
- [Automatic Neural Network Pruning that Efficiently Preserves the Model Accuracy](https://mlsanity.com/paper/2111.09635/) (2021)

## Built on by (most cited first, 30 of 65)

- [Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching](https://mlsanity.com/paper/2406.01733/) (2024)
- [TinyFusion: Diffusion Transformers Learned Shallow](https://mlsanity.com/paper/2412.01199/) (2024)
- [Diversifying the Expert Knowledge for Task-Agnostic Pruning in Sparse Mixture-of-Experts](https://mlsanity.com/paper/2407.09590/) (2024)
- [LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers](https://mlsanity.com/paper/2412.12444/) (2024)
- [Efficient Diffusion Models: A Survey](https://mlsanity.com/paper/2502.06805/) (2025)
- [Sparse-vDiT: Unleashing the Power of Sparse Attention to Accelerate Video Diffusion Transformers](https://mlsanity.com/paper/2506.03065/) (2025)
- [DiTFastAttnV2: Head-wise Attention Compression for Multi-Modality Diffusion Transformers](https://mlsanity.com/paper/2503.22796/) (2025)
- [OBS-Diff: Accurate Pruning For Diffusion Models in One-Shot](https://mlsanity.com/paper/2510.06751/) (2025)
- [FullDiT2: Efficient In-Context Conditioning for Video Diffusion Transformers](https://mlsanity.com/paper/2506.04213/) (2025)
- [Attend to Not Attended: Structure-then-Detail Token Merging for Post-training DiT Acceleration](https://mlsanity.com/paper/2505.11707/) (2025)
- [K-Sort Arena: Efficient and Reliable Benchmarking for Generative Models via K-wise Human Preferences](https://mlsanity.com/paper/2408.14468/) (2024)
- [Not All Prompts Are Made Equal: Prompt-based Pruning of Text-to-Image Diffusion Models](https://mlsanity.com/paper/2406.12042/) (2024)
- [CacheQuant: Comprehensively Accelerated Diffusion Models](https://mlsanity.com/paper/2503.01323/) (2025)
- [Continual Learning for Generative AI: From LLMs to MLLMs and Beyond](https://mlsanity.com/paper/2506.13045/) (2025)
- [Hybrid SD: Edge-Cloud Collaborative Inference for Stable Diffusion Models](https://mlsanity.com/paper/2408.06646/) (2024)
- [BWCache: Accelerating Video Diffusion Transformers through Block-Wise Caching](https://mlsanity.com/paper/2509.13789/) (2025)
- [KIND: Knowledge Integration and Diversion for Training Decomposable Models](https://mlsanity.com/paper/2408.07337/) (2024)
- [Learnable Sparsity for Vision Generative Models](https://mlsanity.com/paper/2412.02852/) (2024)
- [Fast RoPE Attention: Combining the Polynomial Method and Fast Fourier Transform](https://mlsanity.com/paper/2505.11892/) (2025)
- [Scaling Down Text Encoders of Text-to-Image Diffusion Models](https://mlsanity.com/paper/2503.19897/) (2025)
- [HierarchicalPrune: Position-Aware Compression for Large-Scale Diffusion Models](https://mlsanity.com/paper/2508.04663/) (2025)
- [Denoising as Path Planning: Training-Free Acceleration of Diffusion Models with DPCache](https://mlsanity.com/paper/2602.22654/) (2026)
- [Only Large Weights (And Not Skip Connections) Can Prevent the Perils of Rank Collapse](https://mlsanity.com/paper/2505.16284/) (2025)
- [TinySR: Pruning Diffusion for Real-World Image Super-Resolution](https://mlsanity.com/paper/2508.17434/) (2025)
- [Pluggable Pruning with Contiguous Layer Distillation for Diffusion Transformers](https://mlsanity.com/paper/2511.16156/) (2025)
- [SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware Caching](https://mlsanity.com/paper/2602.24208/) (2026)
- [FP4DiT: Towards Effective Floating Point Quantization for Diffusion Transformers](https://mlsanity.com/paper/2503.15465/) (2025)
- [RegionE: Adaptive Region-Aware Generation for Efficient Image Editing](https://mlsanity.com/paper/2510.25590/) (2025)
- [Efficient Pruning of Text-to-Image Models: Insights from Pruning Stable Diffusion](https://mlsanity.com/paper/2411.15113/) (2024)
- [ZipIR: Latent Pyramid Diffusion Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2504.08591/) (2025)

## Abstract (arXiv)

Latent Diffusion Models (LDMs) have emerged as powerful generative models, known for delivering remarkable results under constrained computational resources. However, deploying LDMs on resource-limited devices remains a complex issue, presenting challenges such as memory consumption and inference speed. To address this issue, we introduce LD-Pruner, a novel performance-preserving structured pruning method for compressing LDMs. Traditional pruning methods for deep neural networks are not tailored to the unique characteristics of LDMs, such as the high computational cost of training and the absence of a fast, straightforward and task-agnostic method for evaluating model performance. Our method tackles these challenges by leveraging the latent space during the pruning process, enabling us to effectively quantify the impact of pruning on model performance, independently of the task at hand. This targeted pruning of components with minimal impact on the output allows for faster convergence during training, as the model has less information to re-learn, thereby addressing the high computational cost of training. Consequently, our approach achieves a compressed model that offers improved inference speed and reduced parameter count, while maintaining minimal performance degradation. We demonstrate the effectiveness of our approach on three different tasks: text-to-image (T2I) generation, Unconditional Image Generation (UIG) and Unconditional Audio Generation (UAG). Notably, we reduce the inference time of Stable Diffusion (SD) by 34.9% while simultaneously improving its FID by 5.2% on MS-COCO T2I benchmark. This work paves the way for more efficient pruning methods for LDMs, enhancing their applicability.