awesome-dit-inference
📚A curated list of Awesome Diffusion Inference Papers with Codes: Sampling, Cache, Quantization, Parallelism, etc.🎉
https://github.com/xlite-dev/awesome-dit-inference
Last synced: 2 days ago
JSON representation
-
📙 Sampling
- **DDPM**
- **DDIM**
- **PNDM** - liu/PNDM)  |⭐️⭐️ |
- **DPM-Solver** - solver]](https://github.com/LuChengTHU/dpm-solver)  |⭐️⭐️ |
- **DPM-Solver++** - solver]](https://github.com/LuChengTHU/dpm-solver)  |⭐️⭐️ |
- **DPM-Solver-v3** - Solver-v3]](https://github.com/thu-ml/DPM-Solver-v3)  |⭐️⭐️ |
- **Parallel Sampling**
- **SAMPLER SCHEDULER**
- **Parallel Sampling** - Diffusion]](https://github.com/TZW1998/ParaTAA-Diffusion)  |⭐️⭐️ |
- **YONOS**
- **S^2-DM**
- **StepSaver**
- **DC-Solver** - Solver]](https://github.com/wl-zhao/DC-Solver)  |⭐️⭐️ |
-
📙 Caching
- **ElasticCache-LVLM**
- **Cache-Enabled Sparse Diffusion**
- **DeepCache**
- **Block Caching**
- **Approximate Caching**
- **Layer Caching** - to-cache]](https://github.com/horseee/learning-to-cache/) | ⭐️⭐️ |
- **Fast-Forward Caching(DiT)**
- **Faster I2V Generation**
- **T-GATE V1** - GATE]](https://github.com/HaozheLiu-ST/T-GATE) |⭐️⭐️ |
- **T-GATE V2** - GATE]](https://github.com/HaozheLiu-ST/T-GATE) |⭐️⭐️ |
- **DiTFastAttn** - nics/DiTFastAttn) |⭐️⭐️ |
- **∆-DiT**
- **TokenCache**
- **AdaCache** - DiT/AdaCache) |⭐️⭐️ |
- **TeaCache**
- **LazyDiT**
- **Ca2-VDM** - VDM]](https://github.com/Dawn-LX/CausalCache-VDM/) |⭐️⭐️ |
- **SmoothCache**
- **FasterCache**
- **ToCa** - Z/ToCa) |⭐️⭐️ |
- **SkipCache** - DiT]](https://github.com/OpenSparseLLMs/Skip-DiT) |⭐️⭐️ |
- **DuCa** - Z/DuCa) |⭐️⭐️ |
- **FBCache**
- **FlexCache**
- **Token Pruning** - epiclab/DaTo) |⭐️⭐️ |
- **AB-Cache**
- **DiTFastAttnV2** - nics/DiTFastAttn) |⭐️⭐️ |
- **Increment-Calibrated Cache**
- **FastCache** - xDiT]](https://github.com/NoakLiu/FastCache-xDiT) |⭐️⭐️ |
- **TaylorSeers** - Z/TaylorSeer) |⭐️⭐️ |
- **BACache**
- **DBPrune** - dit]](https://github.com/vipshop/cache-dit) |⭐️⭐️ |
-
📙 Parallelism
- **AsyncDiff**
- **Clip Parallelism** - Infinity]](https://github.com/Yuanshi9815/Video-Infinity) |⭐️⭐️ |
- **FIFO-Diffusion** - Diffusion]](https://github.com/jjihwan/FIFO-Diffusion_public)  |⭐️⭐️ |
- **ParaAttention**
- **PipeFusion** - project/xDiT) | ⭐️⭐️ |
- **TensorRT-LLM SDXL** - TensorRT-LLM]](https://github.com/NVIDIA/TensorRT-LLM/pull/1514) | ⭐️⭐️ |
-
📙 Attention
- **SpargeAttention** - ml/SpargeAttn)  | ⭐️⭐️ |
- **SageAttention-3** - ml/SageAttention)  | ⭐️⭐️ |
- **DraftAttention** - attention]](https://github.com/shawnricecake/draft-attention)  | ⭐️⭐️ |
- **SageAttention** - ml/SageAttention)  | ⭐️⭐️ |
- **SageAttention-2** - ml/SageAttention)  | ⭐️⭐️ |
-
📖 News 🔥🔥
- 2025-07-13 - fast](https://github.com/huggingface/flux-fast) that **makes flux-fast even faster** with **[cache-dit](https://github.com/vipshop/cache-dit)**, **3.3x** speedup on NVIDIA L20 while still maintaining **high precision**.
-
📙 Quantization
- **Transfusion** - pytorch]](https://github.com/A-suozhang/MixDQ) |⭐️⭐️ |
- **MixDQ** - nics/MixDQ) |⭐️⭐️|
- **ViDiT-Q** - nics/ViDiT-Q) |⭐️⭐️|
- **VQ4DiT**
- **LBQ**
- **EE-Diffusion** - diffusion]](https://github.com/taehong-moon/ee-diffusion) |⭐️⭐️ |
- **TFM-PTQ**
- **Diffusion-RWKV** - RWKV]](https://github.com/feizc/Diffusion-RWKV) |⭐️⭐️ |
- **LinFusion**
- **SVDQuant** - han-lab/nunchaku) |⭐️⭐️ |
Programming Languages
Sub Categories