Awesome Efficient Diffusion [](https://awesome.re)

September 19, 2026 · View on GitHub

A collection of papers and code on efficient diffusion and flow matching models for image, video, and 3D generation, world models, and diffusion language models. Topics include fast sampling, distillation, feature and KV caching, parallel decoding, efficient attention, quantization, pruning, model compression, training, and deployment. Contributions are welcome.

Benchmarks · Survey Papers
Papers by Year
2026 · 2025 · 2024 · 2023 · 2022
Implementations · Related Repositories · Contributing

Benchmarks

Resource	What it covers
GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment NeurIPS 2023, Datasets and Benchmarks	Objects, counts, colors, positions, and compositional text-image alignment.
VBench: Comprehensive Benchmark Suite for Video Generative Models CVPR 2024	Multiple dimensions of video generation quality.

Survey Papers

Survey	Venue
Efficient Diffusion Models: A Comprehensive Survey From Principles to Practices Published version	IEEE TPAMI 2025
Efficient Diffusion Models: A Survey	TMLR 2025

Papers by Year

Published papers are listed by venue year; preprints by first release year.

2026

[ICLR] QVGen: Pushing the Limit of Quantized Video Generative Models [code]
[ICLR] Q&C: When Quantization Meets Cache in Efficient Generation
[ICLR] DVD-Quant: Data-free Video Diffusion Transformers Quantization
[ICLR] QuantSparse: Comprehensively Compressing Video Diffusion Transformer with Model Quantization and Attention Sparsification [code]
[ICLR] Quant-dLLM: Post-Training Extreme Low-Bit Quantization for Diffusion Large Language Models [code]
[ICLR] pi-Flow: Policy-Based Few-Step Generation via Imitation Distillation [code]
[ICLR] TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows [code]
[ICLR] Joint Distillation for Fast Likelihood Evaluation and Sampling in Flow-based Models
[ICLR] Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency [code]
[ICLR] Rolling Forcing: Autoregressive Long Video Diffusion in Real Time
[ICLR] SenseFlow: Scaling Distribution Matching for Flow-based Text-to-Image Distillation [code]
[ICLR] DiCache: Let Diffusion Model Determine Its Own Cache
[ICLR] ERTACache: Error Rectification and Timesteps Adjustment for Efficient Diffusion
[ICLR] Relational Feature Caching for Accelerating Diffusion Transformers
[ICLR] ScalingCache: Extreme Acceleration of DiTs through Difference Scaling and Dynamic Interval Caching
[ICLR] Beyond Uniformity: Sample and Frequency Meta Weighting for Post-Training Quantization of Diffusion Models
[ICLR] Gradient-Aligned Calibration for Post-Training Quantization of Diffusion Models
[ICLR] SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer [code]
[ICLR] Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding [code]
[ICLR] Fast-dLLM v2: Efficient Block-Diffusion LLM [code]
[ICLR] d²Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching [code]
[ICLR] FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion [code]
[ICLR] Attention Is All You Need for KV Cache in Diffusion LLMs [code]
[ICLR] Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing [code]
[ICLR] Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct
[ICLR] dParallel: Learnable Parallel Decoding for dLLMs
[ICLR] Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principles
[ICLR] DPad: Efficient Diffusion Language Models with Suffix Dropout
[ICLR] Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes
[ICLR] Diffusion Language Model Knows the Answer Before It Decodes
[ICLR] The Diffusion Duality, Chapter II: Ψ-Samplers and Efficient Curriculum [code]
[ICML] Q-DiT4SR: Exploration of Detail-Preserving Diffusion Transformer Quantization for Real-World Image Super-Resolution [project]
[ICML] RobuQ: Pushing DiTs to W1.58A2 via Robust Activation Quantization [code]
[ICML] WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching [code]
[ICML] Fast-SAM3D: 3Dfy Anything in Images but Faster [code]
[ICML] Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation [code]
[ICML] Light Forcing: Accelerating Autoregressive Video Diffusion via Sparse Attention [code]
[ICML] Veda: Scalable Video Diffusion via Distilled Sparse Attention
[ICML] Deep Forcing: Training-Free Long Video Generation with Deep Sink and Participative Compression
[ICML] FAST-AR: Fast Autoregressive Video Diffusion and World Models with Temporal Cache Compression and Sparse Attention
[ICML] Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization [code]
[ICML] DLLMQuant: A Post-Training Quantization Framework Tailored for Diffusion-Based Large Language Models
[ICML] dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Caching [code]
[ICML] DFlash: Block Diffusion for Flash Speculative Decoding [code]
[CVPR] FlashVSR: Towards Real-time Diffusion-Based Streaming Video Super Resolution [code]
[CVPR] VDOT: Efficient Unified Video Creation via Optimal Transport Distillation [code]
[CVPR] Accelerating Autoregressive Video Diffusion via History-Guided Cache and Residual Correction
[CVPR] D2Cache: Second-Order Delta Caching for Higher Video Diffusion Acceleration
[CVPR] DisCa: Accelerating Video Diffusion Transformers with Distillation-Compatible Learnable Feature Caching [code]
[CVPR] SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware Caching
[CVPR] BinaryAttention: One-Bit QK-Attention for Vision and Diffusion Transformers
[CVPR] DeltaQuant: 4-bit Video Diffusion Models with Spatiotemporal Delta Smoothing
[CVPR] SegQuant: A Semantics-Aware and Generalizable Quantization Framework for Diffusion Models [code]
[CVPR] Sampling-Aware Quantization for Diffusion Models
[CVPR] LinVideo: A Post-Training Framework towards O(n) Attention in Efficient Video Generation
[CVPR] From Sketch to Fresco: Efficient Diffusion Transformer with Progressive Resolution [project]
[CVPR] LESA: Learnable Stage-Aware Predictors for Diffusion Model Acceleration
[CVPR] Beyond Fixed Formulas: Data-Driven Linear Predictor for Efficient Diffusion Models [code]
[CVPR] Forecast the Principal, Stabilize the Residual: Subspace-Aware Feature Caching for Diffusion Transformers
[ECCV] AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation
[ECCV] ResilPhase: Plug-and-Play Phase Mapping and Noise-Resilient Macro-Trajectory Extrapolation for Diffusion Acceleration
[ECCV] DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space [code]
[ECCV] WorldCache: Content-Aware Caching for Accelerated Video World Models [code]
[ECCV] DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion Inference
[ECCV] Accelerating Diffusion Transformers with Gaussian Process Rectified Feature Cache
[AAAI] TR-DQ: Time-Rotation Diffusion Quantization
[AAAI] Sparse-vDiT: Unleashing the Power of Sparse Attention to Accelerate Video Diffusion Transformers
[AAAI] Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Eviction
[MLSys] TiDAR: Think in Diffusion, Talk in Autoregression
[ACL] Focus-dLLM: Accelerating Long-Context Diffusion LLM Inference via Confidence-Guided Context Focusing [project]
[TPAMI] MPQ-DMv2: Flexible Residual Mixed Precision Quantization for Low-Bit Diffusion Models With Temporal Distillation
[arXiv] DSAQuant: Denoising-Stage-Aligned Quantization-Aware Training for Video Generation [code]
[arXiv] Multi-Resolution Flow Matching: Training-Free Diffusion Acceleration via Staged Sampling [code]
[arXiv] Vidu S1: A Real-Time Interactive Video Generation Model [project]
[arXiv] Asymmetric Flow Models [code]
[arXiv] Unlocking Lossless Speedups in LLMs via Discrete Diffusion [code]

2025

[ICLR] BinaryDM: Accurate Weight Binarization for Efficient Diffusion Models [code]
[ICLR] Faster Diffusion Sampling with Randomized Midpoints: Sequential and Parallel
[ICLR] One Step Diffusion via Shortcut Models [code]
[ICLR] Simple ReFlow: Improved Techniques for Fast Flow Models
[ICLR] Simplifying, Stabilizing and Scaling Continuous-Time Consistency Models
[ICLR] Accelerating Diffusion Transformers with Token-wise Feature Caching [code]
[ICLR] Real-Time Video Generation with Pyramid Attention Broadcast
[ICLR] SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Acceleration [code]
[ICLR] DGQ: Distribution-Aware Group Quantization for Text-to-Image Diffusion Models
[ICLR] SVDQuant: Absorbing Outliers by Low-Rank Component for 4-Bit Diffusion Models [code]
[ICLR] ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation [code]
[ICLR] Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models [code]
[ICLR] SANA: Efficient High-Resolution Text-to-Image Synthesis with Linear Diffusion Transformers [code]
[ICLR] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
[ICLR] FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality [code]
[ICLR] Dynamic Diffusion Transformer [code]
[ICLR] Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models [code]
[ICML] Q-VDiT: Towards Accurate Quantization and Distillation of Video-Generation Diffusion Transformers [code]
[ICML] Diffusion Adversarial Post-Training for One-Step Video Generation
[ICML] SpargeAttention: Accurate and Training-free Sparse Attention Accelerating Any Model Inference [code]
[ICML] SADA: Stability-guided Adaptive Diffusion Acceleration [code]
[ICML] Fast Video Generation with Sliding Tile Attention [code]
[ICML] SageAttention2: Efficient Attention with Thorough Outlier Smoothing and Per-thread INT4 Quantization [code]
[ICML] Sparse Video-Gen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity [code]
[ICML] Modulated Diffusion: Accelerating Generative Modeling with Modulated Quantization [code]
[ICML] SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer [code]
[ICML] The Diffusion Duality [code]
[NeurIPS] S²Q-VDiT: Accurate Quantized Video Diffusion Transformer with Salient Data and Sparse Token Distillation [code]
[NeurIPS] Mean Flows for One-step Generative Modeling
[NeurIPS] Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion
[NeurIPS] Shortcutting Pre-trained Flow Matching Diffusion
[NeurIPS] Faster Video Diffusion with Trainable Sparse Attention
[NeurIPS] Radial Attention: O(n log n) Sparse Attention with Energy Decay for Long Video Generation
[NeurIPS] SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-bit Training [code]
[NeurIPS] Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation [code]
[NeurIPS] VETA-DiT: Variance-Equalized and Temporally Adaptive Quantization for Efficient 4-bit Diffusion Transformers
[NeurIPS] PipeFusion: Patch-level Pipeline Parallelism for Diffusion Transformers Inference [code]
[NeurIPS] LaViDa: A Large Diffusion Language Model for Multimodal Understanding [code]
[NeurIPS] Encoder-Decoder Diffusion Language Models for Efficient Training and Inference [code]
[NeurIPS] Fast Solvers for Discrete Diffusion Models: Theory and Applications of High-Order Algorithms [code]
[CVPR] PassionSR: Post-Training Quantization with Adaptive Scale in One-Step Diffusion based Image Super-Resolution [code]
[CVPR] From Slow Bidirectional to Fast Autoregressive Video Diffusion Models
[CVPR] CacheQuant: Comprehensively Accelerated Diffusion Models
[CVPR] Timestep Embedding Tells: It's Time to Cache for Video Diffusion Model [code]
[CVPR] Q-DiT: Accurate Post-Training Quantization for Diffusion Transformers [code]
[ICCV] StreamDiffusion: A Pipeline-level Solution for Real-Time Interactive Generation [code]
[ICCV] Adversarial Distribution Matching for Diffusion Distillation Towards Efficient Image and Video Synthesis
[ICCV] SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation [code]
[ICCV] Adaptive Caching for Faster Video Generation with Diffusion Transformers
[ICCV] From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers
[ICCV] QuantCache: Adaptive Importance-Guided Quantization with Hierarchical Latent and Layer Caching for Video Generation
[ICCV] Training-free and Adaptive Sparse Attention for Efficient Long Video Generation
[ICCV] DMQ: Dissecting Outliers of Diffusion Models for Post-Training Quantization [code]
[ICCV] QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuning [code]
[ICCV] DC-AE 1.5: Accelerating Diffusion Model Convergence with Structured Latent Space [code]
[ICCV] REPA-E: Unlocking VAE for End-to-End Tuning of Latent Diffusion Transformers [code]
[ICCV] Text Embedding Knows How to Quantize Text-Guided Diffusion Models
[ICCV] Memory-Efficient Generative Models via Product Quantization [code]
[AAAI] MPQ-DM: Mixed Precision Quantization for Extremely Low Bit Diffusion Models
[AAAI] Flash Diffusion: Accelerating Any Conditional Diffusion Model for Few Steps Image Generation [code]
[AAAI] D2-DPM: Dual Denoising for Quantized Diffusion Probabilistic Models [code]
[AAAI] Optimizing Quantized Diffusion Models via Distillation with Cross-Timestep Error Correction
[AAAI] Qua2SeDiMo: Quantifiable Quantization Sensitivity of Diffusion Models
[AAAI] TCAQ-DM: Timestep-Channel Adaptive Quantization for Diffusion Models
[ACM MM] DilateQuant: Accurate and Efficient Quantization-Aware Training for Diffusion Models via Weight Dilation
[ACM MM] SpeCa: Accelerating Diffusion Transformers with Speculative Feature Caching [code]
[ACM MM] Compute Only 16 Tokens in One Timestep: Accelerating Diffusion Transformers with Cluster-Driven Feature Caching [code]
[TPAMI] Temporal Feature Matters: A Framework for Diffusion Model Quantization [code]
[Machine Intelligence Research] DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models [code]
[arXiv] Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer [code]
[arXiv] One Small Step in Latent, One Giant Leap for Pixels: Fast Latent Upscale Adapter for Your Diffusion Models [code]
[arXiv] TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times [code]
[arXiv] Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference

2024

[ICLR] Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion
[ICLR] Improved Techniques for Training Consistency Models
[ICLR] InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generation [code]
[ICLR] EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models
[ICLR] PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis
[ICML] Align Your Steps: Optimizing Sampling Schedules in Diffusion Models
[ICML] Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation
[NeurIPS] BiDM: Pushing the Limit of Quantization for Diffusion Models
[NeurIPS] Binarized Diffusion Model for Image Super-Resolution [code]
[NeurIPS] Hyper-SD: Trajectory Segmented Consistency Model for Efficient Image Synthesis
[NeurIPS] Improved Distribution Matching Distillation for Fast Image Synthesis [code]
[NeurIPS] Improving the Training of Rectified Flows
[NeurIPS] Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching [code]
[NeurIPS] DiTFastAttn: Attention Compression for Diffusion Transformer Models
[NeurIPS] BitsFusion: 1.99 bits Weight Quantization of Diffusion Model
[NeurIPS] PTQ4DiT: Post-training Quantization for Diffusion Transformers
[NeurIPS] StepbaQ: Stepping backward as Correction for Quantized Diffusion Models
[NeurIPS] Immiscible Diffusion: Accelerating Diffusion Training with Noise Assignment
[NeurIPS] Simple and Effective Masked Diffusion Language Models [code]
[CVPR] UFOGen: You Forward Once Large Scale Text-to-Image Generation via Diffusion GANs
[CVPR] One-step Diffusion with Distribution Matching Distillation
[CVPR] DeepCache: Accelerating Diffusion Models for Free [code]
[CVPR] TFMQ-DM: Temporal Feature Maintenance Quantization for Diffusion Models
[CVPR] Towards Accurate Post-training Quantization for Diffusion Models
[CVPR] Analyzing and Improving the Training Dynamics of Diffusion Models
[CVPR] DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models
[ECCV] Adversarial Diffusion Distillation
[ECCV] Distilling Diffusion Models into Conditional GANs
[ECCV] Memory-Efficient Fine-Tuning for Quantized Diffusion Model
[ECCV] MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization
[ECCV] Post-training Quantization with Progressive Calibration and Activation Relaxing for Text-to-Image Diffusion Models
[ECCV] Timestep-Aware Correction for Quantized Diffusion Models
[ECCV] BK-SDM: A Lightweight, Fast, and Cheap Version of Stable Diffusion
[ECCV] MobileDiffusion: Instant Text-to-Image Generation on Mobile Devices
[ACM MM] QVD: Post-training Quantization for Video Diffusion Models
[arXiv] Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation
[arXiv] SDXL-Lightning: Progressive Adversarial Diffusion Distillation [models]

2023

[ICLR] Fast Sampling of Diffusion Models with Exponential Integrator [code]
[ICLR] Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow
[ICLR] Flow Matching for Generative Modeling
[ICML] Consistency Models
[NeurIPS] DPM-Solver-v3: Improved Diffusion ODE Solver with Empirical Model Statistics [code]
[NeurIPS] SEEDS: Exponential SDE Solvers for Fast High-Quality Sampling from Diffusion Models
[NeurIPS] UniPC: A Unified Predictor-Corrector Framework for Fast Sampling of Diffusion Models [code]
[NeurIPS] PTQD: Accurate Post-Training Quantization for Diffusion Models [code]
[NeurIPS] Q-DM: An Efficient Low-bit Quantized Diffusion Model
[NeurIPS] Temporal Dynamic Quantization for Diffusion Models
[NeurIPS] Structural Pruning for Diffusion Models [code]
[NeurIPS] SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds
[CVPR] On Distillation of Guided Diffusion Models
[CVPR] Post-training Quantization on Diffusion Models [code]
[CVPR] Multi-Concept Customization of Text-to-Image Diffusion
[CVPR Workshop (ECV)] Token Merging for Fast Stable Diffusion [code]
[ICCV] AutoDiffusion: Training-Free Optimization of Time Steps and Architectures for Automated Diffusion Model Acceleration
[ICCV] Q-diffusion: Quantizing Diffusion Models [code]
[ICCV] DiffFit: Unlocking Transferability of Large Diffusion Models via Simple Parameter-Efficient Fine-Tuning
[ICCV] Efficient Diffusion Training via Min-SNR Weighting Strategy
[ICCV] SVDiff: Compact Parameter Space for Diffusion Fine-Tuning
[EMNLP Findings] DiffuSeq-v2: Bridging Discrete and Continuous Text Spaces for Accelerated Seq2Seq Diffusion Models [code]
[arXiv] LCM-LoRA: A Universal Stable-Diffusion Acceleration Module [code]
[arXiv] Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference

2022

[ICLR] Analytic-DPM: an Analytic Estimate of the Optimal Reverse Variance in Diffusion Probabilistic Models
[ICLR] Learning Fast Samplers for Diffusion Models by Differentiating Through Sample Quality
[ICLR] Pseudo Numerical Methods for Diffusion Models on Manifolds [code]
[ICLR] Progressive Distillation for Fast Sampling of Diffusion Models
[NeurIPS] DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps [code]
[NeurIPS] Elucidating the Design Space of Diffusion-Based Generative Models [code]
[CVPR] High-Resolution Image Synthesis With Latent Diffusion Models [code]

Implementations

xDiT: Parallel inference for diffusion transformers, including PipeFusion.
FastVideo: Video generation inference and post-training, including sparse attention.
Nunchaku: Low-bit diffusion inference and SVDQuant kernels.
SageAttention: Low-precision attention kernels for inference acceleration.
Sparse VideoGen: Sparse attention implementations for video diffusion.
SANA: Efficient image/video model training and inference.
rCM: Continuous-time consistency distillation for video models.
TurboDiffusion: Combined attention acceleration, distillation, and low-precision inference.
Fast-dLLM: KV caching, parallel decoding, and efficient block diffusion for language and multimodal models.
DFlash: Block diffusion drafting for speculative decoding.
Quant-dLLM: Extreme low-bit post-training quantization for diffusion language models.

Awesome Model Quantization: Model quantization across architectures.
Awesome Efficient LLM: Efficient language models.
Efficient Diffusion Models: Resources accompanying the TPAMI survey.

Contributing

Please open a pull request with the paper title, venue, year, paper link, and official code when available. Our scope includes diffusion and flow matching for images, videos, 3D generation, world models, and language, including diffusion-based speculative decoding. We collect relevant publications at leading conferences and journals, as well as recent preprints with early community interest or adoption. For preprints, include a dated source documenting that interest, such as a Hugging Face Daily Papers feature, substantive community discussion, or shared models and integrations. Use the paper's original title without adding acronym suffixes. Keep one entry per paper and update it when the published version becomes available.

Preprint and other inclusion references

Unlocking Lossless Speedups in LLMs via Discrete Diffusion: Released on September 3, 2026; featured as Hugging Face Daily Papers #1 on September 8, with release-week technical discussion and author responses. Evidence
DSAQuant: Released on September 3, 2026, with an official implementation and project page linked from the paper. Implementation
Seed Diffusion: Released on August 4, 2025; release-week Hugging Face discussion on August 6 includes author responses about comparable inference settings and community requests for integration. Evidence
Z-Image: Hugging Face Daily Papers #1 on December 1, 2025, four days after release; the paper page also links community models and Spaces. Evidence
Vidu S1: Hugging Face Daily Papers #1 on July 10, 2026, one week after release. Evidence
Asymmetric Flow Models: Submitted to Hugging Face Daily Papers on May 14, 2026, with release-day models and subsequent community model conversions. Paper and community models
One Small Step in Latent, One Giant Leap for Pixels: Fast Latent Upscale Adapter for Your Diffusion Models: Hugging Face Daily Papers #1 on November 14, 2025, the day after release. Evidence
LCM-LoRA: Hugging Face Daily Papers #1 on November 10, 2023, the day after release, with Diffusers integration documented at launch. Daily Papers Integration
MrFlow: Released on July 2, 2026 and submitted to Hugging Face Daily Papers on July 3; the official repository records release-month Trending Papers coverage and community workflows. Daily Papers Release history and community
Latent Consistency Models: LCM ecosystem integration documented by Hugging Face in November 2023. Evidence
Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation: Hugging Face Daily Papers #1; submitted the day after release. Evidence
TurboDiffusion: Hugging Face Daily Papers #1; submitted within one week of release. Evidence
Token Merging for Fast Stable Diffusion: Release-week Diffusers community benchmarking, March 2023; workshop status is explicit. Evidence
SDXL-Lightning: Release-week public discussion and experimentation, February 2024. Evidence
DPM-Solver++: Author-maintained history records Apple/Hugging Face Swift integration in December 2022; first preprint 2022. Evidence