# Awesome-Diffusion-Inference
**Repository Path**: taipingchanganan/Awesome-Diffusion-Inference
## Basic Information
- **Project Name**: Awesome-Diffusion-Inference
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: GPL-3.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No
## Statistics
- **Stars**: 0
- **Forks**: 0
- **Created**: 2025-02-03
- **Last Updated**: 2025-02-03
## Categories & Tags
**Categories**: Uncategorized
**Tags**: None
## README
📒A curated list of Awesome **Diffusion** Inference Papers with codes. For Awesome LLM Inference, please check 📖[Awesome-LLM-Inference](https://github.com/DefTruth/Awesome-LLM-Inference)  for more details.
## 🤖Contents
- [📙Awesome Diffusion Inference with Sampling](#Sampling)
- [📙Awesome Diffusion Inference with Caching](#Caching)
- [📙Awesome Diffusion Inference with Multi-GPUs](#Distributed)
- [📙Other Awesome Diffusion Inference Paper with codes](#Others)
## ©️Citations
```BibTeX
@misc{Awesome-Diffusion-Inference@2024,
title={Awesome-Diffusion-Inference: A small curated list of Awesome Diffusion Inference with Distributed/Caching/Sampling.},
url={https://github.com/DefTruth/Awesome-Diffusion-Inference},
note={Open-source software available at https://github.com/DefTruth/Awesome-Diffusion-Inference},
author={DefTruth},
year={2024}
}
```
## 📙Awesome Diffusion Inference with Sampling
|Date|Title|Paper|Code|Recom|
|:---:|:---:|:---:|:---:|:---:|
|2020.06| 🔥[**DDPM**] Denoising Diffusion Probabilistic Models(@UC Berkeley) | [[pdf]](https://arxiv.org/abs/2006.11239) |[[diffusion]](https://github.com/hojonathanho/diffusion)  |⭐️⭐️ |
|2020.10| 🔥[**DDIM**] DENOISING DIFFUSION IMPLICIT MODELS(@cs.stanford.edu) | [[pdf]](https://arxiv.org/pdf/2010.02502) |⚠️|⭐️⭐️ |
|2022.02| 🔥[**PNDM**] PSEUDO NUMERICAL METHODS FOR DIFFUSION MODELS ON MANIFOLDS(@) | [[pdf]](https://arxiv.org/pdf/2202.09778) |[[PNDM]](https://github.com/luping-liu/PNDM)  |⭐️⭐️ |
|2022.02| 🔥[**DPM-Solver**] DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps(@Cheng Lu) | [[pdf]](https://arxiv.org/pdf/2206.00927) |[[dpm-solver]](https://github.com/LuChengTHU/dpm-solver)  |⭐️⭐️ |
|2022.11| 🔥[**DPM-Solver++**] DPM-SOLVER++: FAST SOLVER FOR GUIDED SAMPLING OF DIFFUSION PROBABILISTIC MODELS(@Cheng Lu) | [[pdf]](https://arxiv.org/pdf/2211.01095) |[[dpm-solver]](https://github.com/LuChengTHU/dpm-solver)  |⭐️⭐️ |
|2023.10| 🔥[**DPM-Solver-v3**] DPM-Solver-v3: Improved Diffusion ODE Solver with Empirical Model Statistics(@Kaiwen Zheng) | [[pdf]](https://arxiv.org/pdf/2310.13268) |[[DPM-Solver-v3]](https://github.com/thu-ml/DPM-Solver-v3)  |⭐️⭐️ |
|2023.11| 🔥[**Parallel Sampling**] Parallel Sampling of Diffusion Models(@Stanford University) | [[pdf]](https://papers.nips.cc/paper_files/paper/2023/file/0d1986a61e30e5fa408c81216a616e20-Paper-Conference.pdf) | [[paradigms]](https://github.com/AndyShih12/paradigms)  |⭐️⭐️ |
|2023.11| 🔥[**SAMPLER SCHEDULER**] SAMPLER SCHEDULER FOR DIFFUSION MODELS(@sysu)| [[pdf]](https://arxiv.org/pdf/2311.06845) |⚠️|⭐️⭐️ |
|2024.02| 🔥[**Parallel Sampling**] Accelerating Parallel Sampling of Diffusion Models(@Zhiwei Tang) | [[pdf]](https://arxiv.org/pdf/2402.09970) | [[ParaTAA-Diffusion]](https://github.com/TZW1998/ParaTAA-Diffusion)  |⭐️⭐️ |
|2024.01| 🔥[**YONOS**] You Only Need One Step: Fast Super-Resolution with Stable Diffusion via Scale Distillation(@Samsung AI) | [[pdf]](https://arxiv.org/pdf/2401.17258) |⚠️|⭐️⭐️ |
|2024.01| 🔥[**S^2-DM**] S^2-DMs: Skip-Step Diffusion Models(@Yixuan Wang) | [[pdf]](https://arxiv.org/pdf/2401.01520) |⚠️|⭐️⭐️ |
|2024.08| 🔥[**StepSaver**] StepSaver: Predicting Minimum Denoising Steps for Diffusion Model Image Generation(@intel) | [[pdf]](https://arxiv.org/pdf/2408.02054) |⚠️|⭐️⭐️ |
|2024.09| 🔥[**DC-Solver**] DC-Solver: Improving Predictor-Corrector Diffusion Sampler via Dynamic Compensation(@Tsinghua University)| [[pdf]](https://arxiv.org/pdf/2409.03755v1) | [[DC-Solver]](https://github.com/wl-zhao/DC-Solver)  |⭐️⭐️ |
## 📙Awesome Diffusion Inference with Caching
- **UNet Based (DeepCache)**
- **DiT Based (Fast-Forward Caching)**
|Date|Title|Paper|Code|Recom|
|:---:|:---:|:---:|:---:|:---:|
|2023.05|🔥🔥[**Cache-Enabled Sparse Diffusion**] Accelerating Text-to-Image Editing via Cache-Enabled Sparse Diffusion Inference(@pku.edu.cn etc)|[[pdf]](https://arxiv.org/pdf/2305.17423) |⚠️|⭐️⭐️ |
|2023.12|🔥🔥[**DeepCache**] DeepCache: Accelerating Diffusion Models for Free(@nus.edu)|[[pdf]](https://arxiv.org/pdf/2312.00858) | [[DeepCache]](https://github.com/horseee/DeepCache) | ⭐️⭐️ |
|2023.12|🔥🔥[**Block Caching**] Cache Me if You Can: Accelerating Diffusion Models through Block Caching(@Meta GenAI etc)|[[pdf]](https://arxiv.org/pdf/2312.03209) |⚠️|⭐️⭐️ |
|2023.12|🔥🔥[**Approximate Caching**] Approximate Caching for Efficiently Serving Diffusion Models(@Adobe)|[[pdf]](https://arxiv.org/pdf/2312.04429) |⚠️|⭐️⭐️ |
|2024.06| 🔥🔥[**Layer Caching**] Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching(@nus.edu) | [[pdf]](https://arxiv.org/pdf/2406.01733) | [[learning-to-cache]](https://github.com/horseee/learning-to-cache/) | ⭐️⭐️ |
|2024.07|🔥[**ElasticCache-LVLM**] Efficient Inference of Vision Instruction-Following Models with Elastic Cache(@Tsinghua University etc)|[[pdf]](https://arxiv.org/pdf/2407.18121)|[[ElasticCache]](https://github.com/liuzuyan/ElasticCache) |⭐️ |
|2024.07| 🔥🔥[**Fast-Forward Caching(DiT)**] FORA: Fast-Forward Caching in Diffusion Transformer Acceleration(@microsoft.com etc) | [[pdf]](https://arxiv.org/pdf/2407.01425) | [[FORA]](https://github.com/prathebaselva/FORA) |⭐️⭐️ |
|2024.07| 🔥🔥[**Faster I2V Generation**] Faster Image2Video Generation: A Closer Look at CLIP Image Embedding’s Impact on Spatio-Temporal Cross-Attentions(@Ashkan Taghipour etc)|[[pdf]](https://arxiv.org/pdf/2407.19205) |⚠️|⭐️⭐️ |
|2024.04| 🔥🔥[**T-GATE V1**] Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models(@Wentian Zhang etc)|[[pdf]](https://arxiv.org/pdf/2404.02747v1) | [[T-GATE]](https://github.com/HaozheLiu-ST/T-GATE) |⭐️⭐️ |
|2024.04| 🔥🔥[**T-GATE V2**] Faster Diffusion via Temporal Attention Decomposition(@Haozhe Liu etc)|[[pdf]](https://arxiv.org/pdf/2404.02747v2) | [[T-GATE]](https://github.com/HaozheLiu-ST/T-GATE) |⭐️⭐️ |
|2024.06| 🔥🔥[**DiTFastAttn**] DiTFastAttn: Attention Compression for Diffusion Transformer Models(@Zhihang Yuan etc)|[[pdf]](https://arxiv.org/pdf/2406.08552) | [[DiTFastAttn]](https://github.com/thu-nics/DiTFastAttn) |⭐️⭐️ |
|2024.06| 🔥🔥[**∆-DiT**] ∆-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers(@Fudan University)|[[pdf]](https://arxiv.org/pdf/2406.01125) | ⚠️|⭐️⭐️ |
|2024.09| 🔥🔥[**TokenCache**] Token Caching for Diffusion Transformer Acceleration(@Institute of Automation, Chinese Academy of Sciences)|[[pdf]](https://arxiv.org/pdf/2409.18523) | ⚠️|⭐️⭐️ |
|2024.11| 🔥🔥[**AdaCache**] Adaptive Caching for Faster Video Generation with Diffusion Transformers(@Meta)|[[pdf]](https://adacache-dit.github.io/clarity/adacache_meta.pdf) | [[AdaCache]](https://github.com/AdaCache-DiT/AdaCache) |⭐️⭐️ |
|2024.11| 🔥🔥[**TeaCache**] Timestep Embedding Tells: It’s Time to Cache for Video Diffusion Model(@Alibaba)| [[pdf]](https://arxiv.org/pdf/2411.19108) | [[TeaCache]](https://github.com/LiewFeng/TeaCache) |⭐️⭐️ |
|2024.11| 🔥🔥[**LazyDiT**] LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers(@Adobe Research)|[[pdf]](https://arxiv.org/pdf/2412.12444)| ⚠️|⭐️⭐️ |
|2024.11| 🔥🔥[**Ca2-VDM**] Ca2-VDM: Efficient Autoregressive Video Diffusion Model with Causal Generation and Cache Sharing(@ZJU)|[[pdf]](https://arxiv.org/pdf/2411.16375) | [[CausalCache-VDM]](https://github.com/Dawn-LX/CausalCache-VDM/) |⭐️⭐️ |
|2024.11| 🔥🔥[**SmoothCache**] SmoothCache: A Universal Inference Acceleration Technique for Diffusion Transformers(@Roblox) | [[pdf]](https://arxiv.org/pdf/2411.10510) | [[SmoothCache]](https://github.com/Roblox/SmoothCache) |⭐️⭐️ |
|2024.10| 🔥🔥[**FasterCache**] FASTERCACHE: TRAINING-FREE VIDEO DIFFUSION MODEL ACCELERATION WITH HIGH QUALITY(@S-Lab)|[[pdf]](https://arxiv.org/pdf/2410.19355) | [[FasterCache]](https://github.com/Vchitect/FasterCache) |⭐️⭐️ |
|2024.10| 🔥🔥[**ToCa**] ToCa: Accelerating Diffusion Transformers with Token-wise Feature Caching(@SJTU)|[[pdf]](https://arxiv.org/pdf/2410.05317) | [[ToCa]](https://github.com/Shenyi-Z/ToCa) |⭐️⭐️ |
|2024.11| 🔥🔥[**SkipCache**] Accelerating Vision Diffusion Transformers with Skip Branches(@SJTU)|[[pdf]](https://arxiv.org/pdf/2411.17616) | [[Skip-DiT]](https://github.com/OpenSparseLLMs/Skip-DiT) |⭐️⭐️ |
|2024.12| 🔥🔥[**DuCa**] Accelerating Diffusion Transformers with Dual Feature Caching(@SJTU)|[[pdf]](https://arxiv.org/pdf/2412.18911) | [[DuCa]](https://github.com/Shenyi-Z/DuCa) |⭐️⭐️ |
|2025.01| 🔥🔥[**FBCache**] Fastest HunyuanVideo Inference with Context Parallelism and First Block Cache on NVIDIA L20 GPUs(@chengzeyi)| [[docs]](https://github.com/chengzeyi/ParaAttention/blob/main/doc/fastest_hunyuan_video.md) | [[ParaAttention]](https://github.com/chengzeyi/ParaAttention) |⭐️⭐️ |
|2025.01| 🔥🔥[**FlexCache**] FlexCache: Flexible Approximate Cache System for Video Diffusion(@University of Waterloo)| [[pdf]](https://arxiv.org/pdf/2501.04012) | ⚠️|⭐️⭐️ |
|2025.01| 🔥🔥[**Token Pruning**] Token Pruning for Caching Better: 9× Acceleration on Stable Diffusion for Free(@SJTU) | [[pdf]](https://arxiv.org/pdf/2501.00375) | [[DaTo]](https://github.com/EvelynZhang-epiclab/DaTo) |⭐️⭐️ |
## 📙Awesome Diffusion Distributed Inference with Multi-GPUs
- **UNet Based: Displaced Patch parallelism (DistriFusion)**
- **DiT Based: Displaced Patch parallelism (PipeFusion)**
|Date|Title|Paper|Code|Recom|
|:---:|:---:|:---:|:---:|:---:|
|2024.02|🔥🔥[**DistriFusion**] DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models(@MIT etc)|[[pdf]](https://arxiv.org/abs/2402.19481) | [[distrifuser]](https://github.com/mit-han-lab/distrifuser) | ⭐️⭐️ |
|2024.05|🔥🔥[**PipeFusion**] PipeFusion: Displaced Patch Pipeline Parallelism for Inference of Diffusion Transformer Models(@Tencent etc)|[[pdf]](https://arxiv.org/pdf/2405.14430) | [[xDiT]](https://github.com/xdit-project/xDiT) | ⭐️⭐️ |
|2024.06| 🔥🔥[**AsyncDiff**] AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising(@nus.edu) | [[pdf]](https://arxiv.org/pdf/2406.06911) | [[AsyncDiff]](https://github.com/czg1225/AsyncDiff) | ⭐️⭐️ |
|2024.05 | 🔥🔥[**TensorRT-LLM SDXL**] SDXL Distributed Inference with TensorRT-LLM and synchronous comm(@Zars19) | [[pdf]](https://arxiv.org/abs/2402.19481) | [[SDXL-TensorRT-LLM]](https://github.com/NVIDIA/TensorRT-LLM/pull/1514) | ⭐️⭐️ |
|2024.06| 🔥🔥[**Clip Parallelism**] Video-Infinity: Distributed Long Video Generation(@nus.edu)|[[pdf]](https://arxiv.org/pdf/2406.16260) | [[Video-Infinity]](https://github.com/Yuanshi9815/Video-Infinity) |⭐️⭐️ |
|2024.05| 🔥🔥[**FIFO-Diffusion**] FIFO-Diffusion: Generating Infinite Videos from Text without Training(@Seoul National University)|[[pdf]](https://arxiv.org/pdf/2405.11473) | [[FIFO-Diffusion]](https://github.com/jjihwan/FIFO-Diffusion_public)  |⭐️⭐️ |
## 📙Other Awesome Diffusion Inference Paper with codes
|Date|Title|Paper|Code|Recom|
|:---:|:---:|:---:|:---:|:---:|
|2024.08| 🔥[**Transfusion**] Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model(@meta)|[[pdf]](https://www.arxiv.org/pdf/2408.11039) | [[transfusion-pytorch]](https://github.com/lucidrains/transfusion-pytorch) |⭐️⭐️ |
|2024.08| 🔥[**VQ4DiT**] VQ4DiT: Efficient Post-Training Vector Quantization for Diffusion Transformers(@ZJU)|[[pdf]](https://arxiv.org/pdf/2408.17131) |⚠️|⭐️⭐️ |
|2024.08| 🔥[**LBQ**] Low-Bitwidth Floating Point Quantization for Efficient High-Quality Diffusion Models(@toronto.edu)|[[pdf]](https://arxiv.org/pdf/2408.06995) |⚠️|⭐️⭐️ |
|2024.08| 🔥[**EE-Diffusion**] A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models(@KAIST AI)|[[pdf]](https://arxiv.org/pdf/2408.05927) | [[ee-diffusion]](https://github.com/taehong-moon/ee-diffusion) |⭐️⭐️ |
|2024.08| 🔥[**TFM-PTQ**] Temporal Feature Matters: A Framework for Diffusion Model Quantization(@SenseTime)|[[pdf]](https://arxiv.org/pdf/2407.19547) |⚠️|⭐️⭐️ |
|2024.08| 🔥[**Diffusion-RWKV**] Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Models(@Zhengcong Fei)|[[pdf]](https://arxiv.org/pdf/2404.04478) | [[Diffusion-RWKV]](https://github.com/feizc/Diffusion-RWKV) |⭐️⭐️ |
|2024.09| 🔥[**LinFusion**] LINFUSION: 1 GPU, 1 MINUTE, 16K IMAGE(@NUS)|[[pdf]](https://arxiv.org/pdf/2409.02097) | [[LinFusion]](https://github.com/Huage001/LinFusion) |⭐️⭐️ |
|2024.11| 🔥🔥[**SVDQuant**] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models|[[pdf]](https://arxiv.org/pdf/2411.05007) | [[nunchaku]](https://github.com/mit-han-lab/nunchaku) |⭐️⭐️ |
## ©️License
GNU General Public License v3.0
## 🎉Contribute
Welcome to star & submit a PR to this repo!
