cache-dit 1.0.3__py3-none-any.whl → 1.0.14__py3-none-any.whl

{cache_dit-1.0.3.dist-info → cache_dit-1.0.14.dist-info}/METADATA

@@ -1,37 +1,36 @@
 Metadata-Version: 2.4
 Name: cache_dit
-Version: 1.0.3
-Summary: A Unified, Flexible and Training-free Cache Acceleration Framework for 🤗Diffusers.
+Version: 1.0.14
+Summary: A Unified and Flexible Inference Engine with Hybrid Cache Acceleration and Parallelism for 🤗Diffusers.
 Author: DefTruth, vipshop.com, etc.
 Maintainer: DefTruth, vipshop.com, etc
-Project-URL: Repository, https://github.com/vipshop/cache-dit.git
-Project-URL: Homepage, https://github.com/vipshop/cache-dit.git
+Project-URL: Repository, https://github.com/vipshop/cache-dit
+Project-URL: Homepage, https://github.com/vipshop/cache-dit
+Project-URL: GitHub, https://github.com/vipshop/cache-dit
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: packaging
 Requires-Dist: pyyaml
 Requires-Dist: torch>=2.7.1
-Requires-Dist: transformers>=4.55.2
 Requires-Dist: diffusers>=0.35.1
-Requires-Dist: scikit-image
-Requires-Dist: scipy
-Requires-Dist: lpips==0.1.4
-Requires-Dist: torchao>=0.12.0
-Requires-Dist: image-reward
-Provides-Extra: all
+Requires-Dist: transformers>=4.55.2
+Provides-Extra: parallelism
+Requires-Dist: einops>=0.8.1; extra == "parallelism"
+Provides-Extra: quantization
+Requires-Dist: torchao>=0.14.1; extra == "quantization"
+Requires-Dist: bitsandbytes>=0.48.1; extra == "quantization"
 Provides-Extra: metrics
+Requires-Dist: scipy; extra == "metrics"
+Requires-Dist: scikit-image; extra == "metrics"
 Requires-Dist: image-reward; extra == "metrics"
-Requires-Dist: pytorch-fid; extra == "metrics"
 Requires-Dist: lpips==0.1.4; extra == "metrics"
 Provides-Extra: dev
+Requires-Dist: packaging; extra == "dev"
 Requires-Dist: pre-commit; extra == "dev"
 Requires-Dist: pytest<8.0.0,>=7.0.0; extra == "dev"
 Requires-Dist: pytest-html; extra == "dev"
 Requires-Dist: expecttest; extra == "dev"
 Requires-Dist: hypothesis; extra == "dev"
-Requires-Dist: transformers; extra == "dev"
-Requires-Dist: diffusers; extra == "dev"
 Requires-Dist: accelerate; extra == "dev"
 Requires-Dist: peft; extra == "dev"
 Requires-Dist: protobuf; extra == "dev"
@@ -39,40 +38,109 @@ Requires-Dist: sentencepiece; extra == "dev"
 Requires-Dist: opencv-python-headless; extra == "dev"
 Requires-Dist: ftfy; extra == "dev"
 Requires-Dist: scikit-image; extra == "dev"
-Requires-Dist: pytorch-fid; extra == "dev"
+Provides-Extra: all
+Requires-Dist: cache-dit[parallelism]; extra == "all"
+Requires-Dist: cache-dit[quantization]; extra == "all"
+Requires-Dist: cache-dit[metrics]; extra == "all"
 Dynamic: license-file
-Dynamic: provides-extra
-Dynamic: requires-dist
 Dynamic: requires-python
 
-<a href="./README.md">📚English</a> | <a href="./README_CN.md">📚中文阅读 </a> 
-
 <div align="center">
-  <img src=https://github.com/vipshop/cache-dit/raw/main/assets/cache-dit-logo.png height="120">
   <p align="center">
-    A <b>Unified</b>, Flexible and Training-free <b>Cache Acceleration</b> Framework for <b>🤗Diffusers</b> <br>
-    ♥️ Cache Acceleration with <b>One-line</b> Code ~ ♥️
+    <h2 align="center">
+        <img src=https://github.com/vipshop/cache-dit/raw/main/assets/cache-dit-logo.png height="90" align="left">
+        A Unified and Flexible Inference Engine with 🤗🎉<br>Hybrid Cache Acceleration and Parallelism for DiTs<br>
+        <a href="https://pepy.tech/projects/cache-dit"><img src=https://static.pepy.tech/personalized-badge/cache-dit?period=total&units=INTERNATIONAL_SYSTEM&left_color=GRAY&right_color=BLUE&left_text=downloads></a>
+        <img src=https://img.shields.io/github/stars/vipshop/cache-dit.svg?style=dark >
+        <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src=https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg ></a> 
+        <a href="https://hellogithub.com/repository/vipshop/cache-dit" target="_blank"><img src="https://api.hellogithub.com/v1/widgets/recommend.svg?rid=b8b03b3b32a449ea84cfc2b96cd384f3&claim_uid=ofSCbzTmdeQk3FD&theme=small" alt="Featured｜HelloGitHub" /></a> 
+        <img src=https://img.shields.io/badge/Models-30+-hotpink.svg >
+    </h2>
   </p>
-  <div align='center'>
-      <img src=https://img.shields.io/badge/Language-Python-brightgreen.svg >
-      <img src=https://img.shields.io/badge/PRs-welcome-blue.svg >
-      <img src=https://img.shields.io/badge/PyPI-pass-brightgreen.svg >
-      <img src=https://static.pepy.tech/badge/cache-dit >
-      <img src=https://img.shields.io/github/stars/vipshop/cache-dit.svg?style=dark >
-      <img src=https://img.shields.io/badge/Release-v1.0-brightgreen.svg >
-  </div>
-  <p align="center">
-    🎉Now, <b>cache-dit</b> covers almost <b>All</b> Diffusers' <b>DiT</b> Pipelines🎉<br>
-    🔥<a href="./examples/pipeline">Qwen-Image</a> | <a href="./examples/pipeline">Qwen-Image-Edit</a> | <a href="./examples/pipeline">Qwen-Image-Edit-Plus </a> 🔥<br>
+</div>
+
+<!--
+<a href="https://pepy.tech/projects/cache-dit"><img src=https://static.pepy.tech/personalized-badge/cache-dit?period=total&units=INTERNATIONAL_SYSTEM&left_color=GRAY&right_color=GREEN&left_text=downloads></a>
+<a href="https://pypi.org/project/cache-dit/"><img src=https://img.shields.io/pypi/dm/cache-dit.svg ></a> 
+-->
+
+## 🔥Hightlight
+
+We are excited to announce that the **first API-stable version (v1.0.0)** of cache-dit has finally been released!
+**[cache-dit](https://github.com/vipshop/cache-dit)** is a **Unified** and **Flexible** inference engine for 🤗 Diffusers, enabling acceleration with just ♥️**one line**♥️ of code. Key features: **Unified Cache APIs**, **Forward Pattern Matching**, **Automatic Block Adapter**, **DBCache**, **DBPrune**, **Hybrid TaylorSeer Calibrator**, **Hybrid Cache CFG**, **Context Parallelism**, **Tensor Parallelism**, **Torch Compile Compatible** and **🎉SOTA** performance.
+
+```bash
+pip3 install -U cache-dit # pip3 install git+https://github.com/vipshop/cache-dit.git
+```
+You can install the stable release of cache-dit from PyPI, or the latest development version from GitHub. Then try ♥️ Cache Acceleration with just **one line** of code ~ ♥️
+```python
+>>> import cache_dit
+>>> from diffusers import DiffusionPipeline
+>>> pipe = DiffusionPipeline.from_pretrained("Qwen/Qwen-Image") # Can be any diffusion pipeline
+>>> cache_dit.enable_cache(pipe) # One-line code with default cache options.
+>>> output = pipe(...) # Just call the pipe as normal.
+>>> stats = cache_dit.summary(pipe) # Then, get the summary of cache acceleration stats.
+>>> cache_dit.disable_cache(pipe) # Disable cache and run original pipe.
+```
+
+### 📚Core Features
+
+- **[🎉Full 🤗Diffusers Support](./docs/User_Guide.md#supported-pipelines)**: Notably, **[cache-dit](https://github.com/vipshop/cache-dit)** now supports nearly **all** of Diffusers' **DiT-based** pipelines, include **[30+](./examples/pipeline/)** series, nearly **[100+](./examples/pipeline/)** pipelines, such as FLUX.1, Qwen-Image, Qwen-Image-Lightning, Wan 2.1/2.2, HunyuanImage-2.1, HunyuanVideo, HiDream, AuraFlow, CogView3Plus, CogView4, CogVideoX, LTXVideo, ConsisID, SkyReelsV2, VisualCloze, PixArt, Chroma, Mochi, SD 3.5, DiT-XL, etc.  
+- **[🎉Extremely Easy to Use](./docs/User_Guide.md#unified-cache-apis)**: In most cases, you only need **one line** of code: `cache_dit.enable_cache(...)`. After calling this API, just use the pipeline as normal.   
+- **[🎉Easy New Model Integration](./docs/User_Guide.md#automatic-block-adapter)**: Features like **Unified Cache APIs**, **Forward Pattern Matching**, **Automatic Block Adapter**, **Hybrid Forward Pattern**, and **Patch Functor** make it highly functional and flexible. For example, we achieved 🎉 Day 1 support for [HunyuanImage-2.1](https://github.com/Tencent-Hunyuan/HunyuanImage-2.1) with 1.7x speedup w/o precision loss—even before it was available in the Diffusers library.  
+- **[🎉State-of-the-Art Performance](./bench/)**: Compared with algorithms including Δ-DiT, Chipmunk, FORA, DuCa, TaylorSeer and FoCa, cache-dit achieved the **SOTA** performance w/ **7.4x↑🎉** speedup on ClipScore!
+- **[🎉Support for 4/8-Steps Distilled Models](./bench/)**: Surprisingly, cache-dit's **DBCache** works for extremely few-step distilled models—something many other methods fail to do.  
+- **[🎉Compatibility with Other Optimizations](./docs/User_Guide.md#️torch-compile)**: Designed to work seamlessly with torch.compile, Quantization ([torchao](./examples/quantize/), [🔥nunchaku](./examples/quantize/)), CPU or Sequential Offloading, **[🔥Context Parallelism](./docs/User_Guide.md/#️hybrid-context-parallelism)**, **[🔥Tensor Parallelism](./docs/User_Guide.md#️hybrid-tensor-parallelism)**, etc.  
+- **[🎉Hybrid Cache Acceleration](./docs/User_Guide.md#taylorseer-calibrator)**: Now supports hybrid **Block-wise Cache + Calibrator** schemes (e.g., DBCache or DBPrune + TaylorSeerCalibrator). DBCache or DBPrune acts as the **Indicator** to decide *when* to cache, while the Calibrator decides *how* to cache. More mainstream cache acceleration algorithms (e.g., FoCa) will be supported in the future, along with additional benchmarks—stay tuned for updates!  
+- **[🤗Diffusers Ecosystem Integration](https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit)**: 🔥**cache-dit** has joined the Diffusers community ecosystem as the **first** DiT-specific cache acceleration framework! Check out the documentation here: <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src=https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg ></a>
+
+![](https://github.com/vipshop/cache-dit/raw/main/assets/clip-score-bench.png)
+
+## 🔥 Supported DiTs
+
+> [!Tip] 
+> One **Model Series** may contain **many** pipelines. cache-dit applies optimizations at the **Transformer** level; thus, any pipelines that include the supported transformer are already supported by cache-dit. ✅: known work and official supported now; ✖️: unofficial supported now, but maybe support in the future; **4-bits**: w/ nunchaku + svdq int4.
+
+<div align="center">
+
+| 📚Model | Cache  | CP | TP | 📚Model | Cache  | CP | TP |
+|:---|:---|:---|:---|:---|:---|:---|:---|
+| **🎉[FLUX.1](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ | **🎉[FLUX.1 4-bits](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ |
+| **🎉[Qwen-Image](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ | **🎉[Qwen-Image 4-bits](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ |
+| **🎉[Qwen...Lightning](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ | **🎉[Qwen...Lightning 4-bits](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ |
+| **🎉[CogVideoX](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ | **🎉[OmniGen](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[Wan 2.1](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ | **🎉[PixArt](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ |
+| **🎉[Wan 2.2](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ | **🎉[CogVideoX 1.5](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ |
+| **🎉[HunyuanVideo](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ | **🎉[Sana](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[LTX](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ | **🎉[VisualCloze](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ |
+| **🎉[Allegro](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[AuraFlow](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[CogView4](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ | **🎉[ShapE](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[CogView3Plus](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ | **🎉[Chroma](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ️✅ |
+| **🎉[Cosmos](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[HiDream](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[EasyAnimate](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[HunyuanDiT](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[SkyReelsV2](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[HunyuanDiTPAG](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[StableDiffusion3](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[Kandinsky5](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✅️ |
+| **🎉[ConsisID](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ | **🎉[PRX](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[DiT](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✖️ | **🎉[HunyuanImage](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✅ | ✅ |
+| **🎉[Amused](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[LongCatVideo](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[StableAudio](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[Bria](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+| **🎉[Mochi](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ | **🎉[Lumina](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline)** | ✅ | ✖️ | ✖️ |
+
+</div>
+
+<details align='center'>
+<summary>🔥<b>Click</b> here to show many <b>Image/Video</b> cases🔥</summary>
+  
+<p align='center'>
+  🎉Now, cache-dit covers almost All Diffusers' DiT Pipelines🎉 <br>
+   🔥<a href="./examples/pipeline">Qwen-Image</a> | <a href="./examples/pipeline">Qwen-Image-Edit</a> | <a href="./examples/pipeline">Qwen-Image-Edit-Plus </a> 🔥<br>
     🔥<a href="./examples/pipeline">FLUX.1</a> | <a href="./examples/pipeline">Qwen-Image-Lightning 4/8 Steps</a> | <a href="./examples/pipeline"> Wan 2.1 </a> | <a href="./examples/pipeline"> Wan 2.2 </a>🔥<br>
     🔥<a href="./examples/pipeline">HunyuanImage-2.1</a> | <a href="./examples/pipeline">HunyuanVideo</a> | <a href="./examples/pipeline">HunyuanDiT</a> | <a href="./examples/pipeline">HiDream</a> | <a href="./examples/pipeline">AuraFlow</a>🔥<br>
     🔥<a href="./examples/pipeline">CogView3Plus</a> | <a href="./examples/pipeline">CogView4</a> | <a href="./examples/pipeline">LTXVideo</a> | <a href="./examples/pipeline">CogVideoX</a> | <a href="./examples/">CogVideoX 1.5</a> | <a href="./examples/">ConsisID</a>🔥<br>
     🔥<a href="./examples/pipeline">Cosmos</a> | <a href="./examples/pipeline">SkyReelsV2</a> | <a href="./examples/pipeline">VisualCloze</a> | <a href="./examples/pipeline">OmniGen 1/2</a> | <a href="./examples/pipeline">Lumina 1/2</a> | <a href="./examples/pipeline">PixArt</a>🔥<br>
     🔥<a href="./examples/pipeline">Chroma</a> | <a href="./examples/pipeline">Sana</a> | <a href="./examples/pipeline">Allegro</a> | <a href="./examples/pipeline">Mochi</a> | <a href="./examples/pipeline">SD 3/3.5</a> | <a href="./examples/pipeline">Amused</a> | <a href="./examples/pipeline"> ... </a> | <a href="./examples/pipeline">DiT-XL</a>🔥
-    <br>♥️ Please consider to leave a <b>⭐️ Star</b> to support us ~ ♥️ 
-  </p>
-</div>
-
+</p>
+  
 <div align='center'>
   <img src=https://github.com/vipshop/cache-dit/raw/main/assets/gifs/wan2.2.C0_Q0_NONE.gif width=124px>
   <img src=https://github.com/vipshop/cache-dit/raw/main/assets/gifs/wan2.2.C1_Q0_DBCACHE_F1B0_W2M8MC2_T1O2_R0.08.gif width=124px>
@@ -95,10 +163,6 @@ Dynamic: requires-python
   <img src=https://github.com/vipshop/cache-dit/raw/main/assets/qwen-image-edit.C0_L0_Q0_DBCACHE_F1B0_W8M0MC2_T0O2_R0.12_S24.png width=125px>
   <p><b>🔥Qwen-Image-Edit</b> | Input w/o Edit | Baseline | <a href="https://github.com/vipshop/cache-dit">+cache-dit</a>:1.6x↑🎉 | 1.9x↑🎉 </p>
 </div>
-
-<details align='center'>
-<summary>🔥<b>Click</b> here to show many <b>Image/Video</b> cases🔥</summary>
-
 <div align='center'>
   <img src=https://github.com/vipshop/cache-dit/raw/main/assets/flux-kontext-cat.C0_L0_Q0_NONE.png width=100px>
   <img src=https://github.com/vipshop/cache-dit/raw/main/assets/flux-kontext.C0_L0_Q0_NONE.png width=100px>
@@ -168,87 +232,35 @@ Dynamic: requires-python
 
 </details>
 
-## 🔥Hightlight <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src=https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg ></a>
-
-We are excited to announce that the **first API-stable version (v1.0.0)** of cache-dit has finally been released!
-**[cache-dit](https://github.com/vipshop/cache-dit)** is a **Unified**, **Flexible**, and **Training-free** cache acceleration framework for 🤗 Diffusers, enabling cache acceleration with just **one line** of code. Key features include **Unified Cache APIs**, **Forward Pattern Matching**, **Automatic Block Adapter**, **Hybrid Forward Pattern**, **DBCache**, **TaylorSeer Calibrator**, and **Cache CFG**.   
-
-```bash
-pip3 install -U cache-dit # pip3 install git+https://github.com/vipshop/cache-dit.git
-```
-You can install the stable release of cache-dit from PyPI, or the latest development version from GitHub. Then try ♥️ Cache Acceleration with just **one line** of code ~ ♥️
-```python
->>> import cache_dit
->>> from diffusers import DiffusionPipeline
->>> pipe = DiffusionPipeline.from_pretrained("Qwen/Qwen-Image") # Can be any diffusion pipeline
->>> cache_dit.enable_cache(pipe) # One-line code with default cache options.
->>> output = pipe(...) # Just call the pipe as normal.
->>> stats = cache_dit.summary(pipe) # Then, get the summary of cache acceleration stats.
->>> cache_dit.disable_cache(pipe) # Disable cache and run original pipe.
-```
-
-### 📚Core Features
-
-- **[🎉Full 🤗Diffusers Support](./docs/User_Guide.md#supported-pipelines)**: Notably, **[cache-dit](https://github.com/vipshop/cache-dit)** now supports nearly **all** of Diffusers' **DiT-based** pipelines, such as Qwen-Image, FLUX.1, Qwen-Image-Lightning, HunyuanImage-2.1, HunyuanVideo, HunyuanDiT, Wan 2.1/2.2, HiDream, AuraFlow, CogView3Plus, CogView4, LTXVideo, CogVideoX 1.5, ConsisID, SkyReelsV2, VisualCloze, OmniGen, Lumina, PixArt, Chroma, Sana, Allegro, Mochi, SD 3.5, Amused, and DiT-XL.  
-- **[🎉Extremely Easy to Use](./docs/User_Guide.md#unified-cache-apis)**: In most cases, you only need **one line** of code: `cache_dit.enable_cache(...)`. After calling this API, just use the pipeline as normal.   
-- **[🎉Easy New Model Integration](./docs/User_Guide.md#automatic-block-adapter)**: Features like **Unified Cache APIs**, **Forward Pattern Matching**, **Automatic Block Adapter**, **Hybrid Forward Pattern**, and **Patch Functor** make it highly functional and flexible. For example, we achieved 🎉 Day 1 support for [HunyuanImage-2.1](https://github.com/Tencent-Hunyuan/HunyuanImage-2.1) with 1.7x speedup w/o precision loss—even before it was available in the Diffusers library.  
-- **[🎉State-of-the-Art Performance](./bench/)**: Compared with algorithms including Δ-DiT, Chipmunk, FORA, DuCa, TaylorSeer and FoCa, cache-dit achieves the best accuracy when the speedup ratio is below 4x.  
-- **[🎉Support for 4/8-Steps Distilled Models](./bench/)**: Surprisingly, cache-dit's **DBCache** works for extremely few-step distilled models—something many other methods fail to do.  
-- **[🎉Compatibility with Other Optimizations](./docs/User_Guide.md#️torch-compile)**: Designed to work seamlessly with torch.compile, model CPU offload, sequential CPU offload, group offloading, etc.  
-- **[🎉Hybrid Cache Acceleration](./docs/User_Guide.md#taylorseer-calibrator)**: Now supports hybrid **DBCache + Calibrator** schemes (e.g., DBCache + TaylorSeerCalibrator). DBCache acts as the **Indicator** to decide *when* to cache, while the Calibrator decides *how* to cache. More mainstream cache acceleration algorithms (e.g., FoCa) will be supported in the future, along with additional benchmarks—stay tuned for updates!  
-- **[🤗Diffusers Ecosystem Integration](https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit)**: 🔥**cache-dit** has joined the Diffusers community ecosystem as the **first** DiT-specific cache acceleration framework! Check out the documentation here: <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src=https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg ></a>
-
-![image-reward-bench](https://github.com/vipshop/cache-dit/raw/main/assets/image-reward-bench.png)
-
-## 🔥Important News
-
-- 2025.10.10: 🔥[**Qwen-Image-ControlNet-Inpainting**](https://huggingface.co/InstantX/Qwen-Image-ControlNet-Inpainting) **2.3x↑🎉** speedup! Check the [example](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_qwen_image_controlnet_inpaint.py).
-- 2025.09.26: 🔥[**Qwen-Image-Edit-Plus(2509)**](https://github.com/QwenLM/Qwen-Image) **2.1x↑🎉** speedup! Please check the [example](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_qwen_image_edit_plus.py).
-- 2025.09.25: 🎉The **first API-stable version (v1.0.0)** of cache-dit has finally been released!
-- 2025.09.25: 🔥**cache-dit** has joined the Diffusers community ecosystem: <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src=https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg ></a>
-- 2025.09.10: 🎉Day 1 support [**HunyuanImage-2.1**](https://github.com/Tencent-Hunyuan/HunyuanImage-2.1) with **1.7x↑🎉** speedup! Check this [example](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_hunyuan_image_2.1.py).
-- 2025.09.08: 🔥[**Qwen-Image-Lightning**](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_qwen_image_lightning.py) **7.1/3.5 steps🎉** inference with **[DBCache: F16B16](https://github.com/vipshop/cache-dit)**.
-- 2025.09.03: 🎉[**Wan2.2-MoE**](https://github.com/Wan-Video) **2.4x↑🎉** speedup! Please refer to [run_wan_2.2.py](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_wan_2.2.py) as an example.
-- 2025.08.19: 🔥[**Qwen-Image-Edit**](https://github.com/QwenLM/Qwen-Image) **2x↑🎉** speedup! Check the example: [run_qwen_image_edit.py](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_qwen_image_edit.py).
-- 2025.08.11: 🔥[**Qwen-Image**](https://github.com/QwenLM/Qwen-Image) **1.8x↑🎉** speedup! Please refer to [run_qwen_image.py](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_qwen_image.py) as an example.
-
-<details>
-<summary>Previous News</summary>
-
-- 2025.09.08: 🎉First caching mechanism in [Wan2.2](https://github.com/Wan-Video/Wan2.2) with **[cache-dit](https://github.com/vipshop/cache-dit)**, check this [PR](https://github.com/Wan-Video/Wan2.2/pull/127) for more details.
-- 2025.09.08: 🎉First caching mechanism in [Qwen-Image-Lightning](https://github.com/ModelTC/Qwen-Image-Lightning) with **[cache-dit](https://github.com/vipshop/cache-dit)**, check this [PR](https://github.com/ModelTC/Qwen-Image-Lightning/pull/35).
-- 2025.08.10: 🔥[**FLUX.1-Kontext-dev**](https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev) is supported! Please refer [run_flux_kontext.py](https://github.com/vipshop/cache-dit/blob/main/examples/pipeline/run_flux_kontext.py) as an example.
-- 2025.08.12: 🎉First caching mechanism in [QwenLM/Qwen-Image](https://github.com/QwenLM/Qwen-Image) with **[cache-dit](https://github.com/vipshop/cache-dit)**, check this [PR](https://github.com/QwenLM/Qwen-Image/pull/61).
-- 2025.07.18: 🎉First caching mechanism in [🤗huggingface/flux-fast](https://github.com/huggingface/flux-fast) with **[cache-dit](https://github.com/vipshop/cache-dit)**, check the [PR](https://github.com/huggingface/flux-fast/pull/13).
-- 2025.07.13: 🎉[**FLUX.1-dev**](https://github.com/xlite-dev/flux-faster) **3.3x↑🎉** speedup! NVIDIA L20 with **[cache-dit](https://github.com/vipshop/cache-dit)** + **compile + FP8 DQ**.
-
-</details>
-
-## 📚User Guide
+## 📖Table of Contents
 
 <div id="user-guide"></div>
 
-For more advanced features such as **Unified Cache APIs**, **Forward Pattern Matching**, **Automatic Block Adapter**, **Hybrid Forward Pattern**, **Patch Functor**, **DBCache**, **TaylorSeer Calibrator**, and **Hybrid Cache CFG**, please refer to the [🎉User_Guide.md](./docs/User_Guide.md) for details.
+For more advanced features such as **Unified Cache APIs**, **Forward Pattern Matching**, **Automatic Block Adapter**, **Hybrid Forward Pattern**, **Patch Functor**, **DBCache**, **DBPrune**, **TaylorSeer Calibrator**, **Hybrid Cache CFG**, **Context Parallelism** and **Tensor Parallelism**, please refer to the [🎉User_Guide.md](./docs/User_Guide.md) for details.
 
 - [⚙️Installation](./docs/User_Guide.md#️installation)
+- [🔥Supported DiTs](./docs/User_Guide.md#supported)
 - [🔥Benchmarks](./docs/User_Guide.md#benchmarks)
-- [🔥Supported Pipelines](./docs/User_Guide.md#supported-pipelines)
 - [🎉Unified Cache APIs](./docs/User_Guide.md#unified-cache-apis)
   - [📚Forward Pattern Matching](./docs/User_Guide.md#forward-pattern-matching)
   - [📚Cache with One-line Code](./docs/User_Guide.md#%EF%B8%8Fcache-acceleration-with-one-line-code)
   - [🔥Automatic Block Adapter](./docs/User_Guide.md#automatic-block-adapter)
-  - [📚Hybird Forward Pattern](./docs/User_Guide.md#hybird-forward-pattern)
+  - [📚Hybrid Forward Pattern](./docs/User_Guide.md#hybrid-forward-pattern)
   - [📚Implement Patch Functor](./docs/User_Guide.md#implement-patch-functor)
+  - [📚Transformer-Only Interface](./docs/User_Guide.md#transformer-only-interface)
+  - [📚How to use ParamsModifier](./docs/User_Guide.md#how-to-use-paramsmodifier)
   - [🤖Cache Acceleration Stats](./docs/User_Guide.md#cache-acceleration-stats-summary)
-- [⚡️Dual Block Cache](./docs/User_Guide.md#️dbcache-dual-block-cache)
-- [🔥TaylorSeer Calibrator](./docs/User_Guide.md#taylorseer-calibrator)
+- [⚡️DBCache: Dual Block Cache](./docs/User_Guide.md#️dbcache-dual-block-cache)
+- [⚡️DBPrune: Dynamic Block Prune](./docs/User_Guide.md#️dbprune-dynamic-block-prune)
 - [⚡️Hybrid Cache CFG](./docs/User_Guide.md#️hybrid-cache-cfg)
-- [🛠Metrics CLI](./docs/User_Guide.md#metrics-cli)
+- [🔥Hybrid TaylorSeer Calibrator](./docs/User_Guide.md#taylorseer-calibrator)
+- [⚡️Hybrid Context Parallelism](./docs/User_Guide.md#context-parallelism)
+- [⚡️Hybrid Tensor Parallelism](./docs/User_Guide.md#tensor-parallelism)
+- [🤖Low-bits Quantization](./docs/User_Guide.md#quantization)
+- [🛠Metrics Command Line](./docs/User_Guide.md#metrics-cli)
 - [⚙️Torch Compile](./docs/User_Guide.md#️torch-compile)
 - [📚API Documents](./docs/User_Guide.md#api-documentation)
 
-
-
 ## 👋Contribute 
 <div id="contribute"></div>
 
@@ -267,15 +279,10 @@ How to contribute? Star ⭐️ this repo to support us or check [CONTRIBUTE.md](
 
 ## 🎉Projects Using CacheDiT
 
-Here is a curated list of open-source projects integrating **CacheDiT**, including popular repositories like [jetson-containers](https://github.com/dusty-nv/jetson-containers/blob/master/packages/diffusion/cache_edit/build.sh) ![](https://img.shields.io/github/stars/dusty-nv/jetson-containers.svg), [flux-fast](https://github.com/huggingface/flux-fast) ![](https://img.shields.io/github/stars/huggingface/flux-fast.svg), and [sdnext](https://github.com/vladmandic/sdnext/blob/dev/modules/cachedit.py) ![](https://img.shields.io/github/stars/vladmandic/sdnext.svg). **CacheDiT** has also been **recommended** by [Wan2.2](https://github.com/Wan-Video/Wan2.2) ![](https://img.shields.io/github/stars/Wan-Video/Wan2.2.svg), [Qwen-Image-Lightning](https://github.com/ModelTC/Qwen-Image-Lightning) ![](https://img.shields.io/github/stars/ModelTC/Qwen-Image-Lightning.svg), [Qwen-Image](https://github.com/QwenLM/Qwen-Image) ![](https://img.shields.io/github/stars/QwenLM/Qwen-Image.svg), and <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src="https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg"></a> ![](https://img.shields.io/github/stars/huggingface/diffusers.svg), among others. We would be grateful if you could let us know if you have used CacheDiT.
-
-## ©️Acknowledgements
+Here is a curated list of open-source projects integrating **CacheDiT**, including popular repositories like [jetson-containers](https://github.com/dusty-nv/jetson-containers/blob/master/packages/diffusion/cache_edit/build.sh) ![](https://img.shields.io/github/stars/dusty-nv/jetson-containers.svg), [flux-fast](https://github.com/huggingface/flux-fast) ![](https://img.shields.io/github/stars/huggingface/flux-fast.svg), and [sdnext](https://github.com/vladmandic/sdnext/discussions/4269) ![](https://img.shields.io/github/stars/vladmandic/sdnext.svg). 🎉**CacheDiT** has been **recommended** by: [Wan2.2](https://github.com/Wan-Video/Wan2.2) ![](https://img.shields.io/github/stars/Wan-Video/Wan2.2.svg), [Qwen-Image-Lightning](https://github.com/ModelTC/Qwen-Image-Lightning) ![](https://img.shields.io/github/stars/ModelTC/Qwen-Image-Lightning.svg), [Qwen-Image](https://github.com/QwenLM/Qwen-Image) ![](https://img.shields.io/github/stars/QwenLM/Qwen-Image.svg), [LongCat-Video](https://github.com/meituan-longcat/LongCat-Video) ![](https://img.shields.io/github/stars/meituan-longcat/LongCat-Video.svg), [Kandinsky-5](https://github.com/ai-forever/Kandinsky-5) ![](https://img.shields.io/github/stars/ai-forever/Kandinsky-5.svg), <a href="https://huggingface.co/docs/diffusers/main/en/optimization/cache_dit"><img src="https://img.shields.io/badge/🤗Diffusers-ecosystem-yellow.svg"></a> ![](https://img.shields.io/github/stars/huggingface/diffusers.svg), [<a href="https://hellogithub.com/repository/vipshop/cache-dit" target="_blank"><img src="https://api.hellogithub.com/v1/widgets/recommend.svg?rid=b8b03b3b32a449ea84cfc2b96cd384f3&claim_uid=ofSCbzTmdeQk3FD&theme=small" alt="Featured｜HelloGitHub" /></a>](https://hellogithub.com/repository/vipshop/cache-dit) ![](https://img.shields.io/github/stars/521xueweihan/HelloGitHub.svg), among others.
 
-<div id="Acknowledgements"></div>
 
-The **cache-dit** codebase is adapted from FBCache. Over time its codebase diverged a lot, and **cache-dit** API is no longer compatible with FBCache. 
-
-## ©️Special Acknowledgements
+## ©️Acknowledgements
 
 Special thanks to vipshop's Computer Vision AI Team for supporting document, testing and production-level deployment of this project.
 
@@ -285,10 +292,10 @@ Special thanks to vipshop's Computer Vision AI Team for supporting document, tes
 
 ```BibTeX
 @misc{cache-dit@2025,
-  title={cache-dit: A Unified, Flexible and Training-free Cache Acceleration Framework for Diffusers.},
+  title={cache-dit: A Unified and Flexible Inference Engine with Hybrid Cache Acceleration and Parallelism for Diffusers.},
   url={https://github.com/vipshop/cache-dit.git},
   note={Open-source software available at https://github.com/vipshop/cache-dit.git},
-  author={vipshop.com},
+  author={DefTruth, vipshop.com},
   year={2025}
 }
 ```

cache_dit-1.0.14.dist-info/RECORD

@@ -0,0 +1,102 @@
+cache_dit/__init__.py,sha256=o2p2odulRXB24ZZb4zE0Rn1O085lRu59sO7BEVCPOOs,2025
+cache_dit/_version.py,sha256=6M51k-eDUkvbg1O0UDaNPjBRHmqrs5f127zkdrldxjg,706
+cache_dit/logger.py,sha256=0zsu42hN-3-rgGC_C29ms1IvVpV4_b4_SwJCKSenxBE,4304
+cache_dit/summary.py,sha256=ZSmmpEEV_yKT6d86rqEIwaQNni1WgOEVdbg1o88-7lI,22159
+cache_dit/utils.py,sha256=FeprLDwxnlWWMqBS5ZKRgqJrORkCn64oTqRVNjHYLJM,1846
+cache_dit/caching/.gitignore,sha256=5Cb-qT9wsTUoMJ7vACDF7ZcLpAXhi5v-xdcWSRit988,23
+cache_dit/caching/__init__.py,sha256=47Im7unxC6gB2vFjFgbY-qHuJ-GOPKT6nU3ZkTGUA74,1747
+cache_dit/caching/cache_interface.py,sha256=OWMv6K8jpSFNc9kjdEim_XdFTn0x5ygUKfM-S_pOtZ8,18058
+cache_dit/caching/cache_types.py,sha256=QnWfaS52UOXQtnoCUOwwz4ziY0dyBta6vQ6hvgtdV44,1404
+cache_dit/caching/forward_pattern.py,sha256=DAnldDC-B_FOMK8REtX2hx8mLZ9GLe1UWYfkxzyyMgo,2197
+cache_dit/caching/params_modifier.py,sha256=4GDKkZmol53j9ICZnbbkuTSAjO2p63KFH3hTE_fahBM,3588
+cache_dit/caching/utils.py,sha256=0K2DSvbhLJgajI9tHtwjaYH_YKAUWK8PY74wPiLRC_s,2473
+cache_dit/caching/block_adapters/__init__.py,sha256=duiVXQB8vasnC15PdGc9yTYQvi-NOjmb3MstmjpbX50,22254
+cache_dit/caching/block_adapters/block_adapters.py,sha256=H5BuIm5_1Cy2aHjSu56WQhrB2VozUfeARiWdMpxBHSk,25267
+cache_dit/caching/block_adapters/block_registers.py,sha256=L-PYmh2jz-D8b-rBznm2lC0k6u9Tpo6u68lXJXM86Cw,3821
+cache_dit/caching/cache_adapters/__init__.py,sha256=IRO0IigSRxxwLccCBNLCJuQMtZDZo9WUikFAtn8LUFE,73
+cache_dit/caching/cache_adapters/cache_adapter.py,sha256=1xeFMybrlSQycw_Slq9rZ2RcwJXkqDvvPp3DpST44Wg,25654
+cache_dit/caching/cache_blocks/__init__.py,sha256=FKKGZ_jfFadet4k9jXejCqp1dhS7sIHBGfyWBtlPeXU,9924
+cache_dit/caching/cache_blocks/offload_utils.py,sha256=wusgcqaCrwEjvv7Guy-6VXhNOgPPUrBV2sSVuRmGuvo,3513
+cache_dit/caching/cache_blocks/pattern_0_1_2.py,sha256=bzSVkxFa2uvtNU47GC-poWR6l74i597hIhXjX9u6AGU,654
+cache_dit/caching/cache_blocks/pattern_3_4_5.py,sha256=g-4WStwVJIRF4gj_71UX6Zt0UqHKv42K-Zf4f8cw6Pc,19767
+cache_dit/caching/cache_blocks/pattern_base.py,sha256=ZaLeyFZDTE1qyV-n6bZNLV50-KPxMNxbEYjhAl7VuTQ,27832
+cache_dit/caching/cache_blocks/pattern_utils.py,sha256=AEuWRQxbR2p6yIUwblAT_kQokqIBgbCEjnJXjyT6YMA,3106
+cache_dit/caching/cache_contexts/__init__.py,sha256=OOiCta6yCWr7NIwj45_ERY8SFdDS_m4usubHM4Pgq8A,805
+cache_dit/caching/cache_contexts/cache_config.py,sha256=7lKrCg6v9scWTV59kulfIYuJkdHYTRw9jhiBlfLdN9w,5947
+cache_dit/caching/cache_contexts/cache_context.py,sha256=kHLAyWK9-jC8Up_y6fgcdJR_NU92L6QQ3oeBs8I8Cu4,11406
+cache_dit/caching/cache_contexts/cache_manager.py,sha256=L7hQbyEqf5TZMJ9aCP81NxDlNHs4ASf8XEZ-09DcVH4,35386
+cache_dit/caching/cache_contexts/context_manager.py,sha256=7Q3D5eQperAWaEuVrlwkCMDSCjqdgzSOtSta1aJ0ZYM,1059
+cache_dit/caching/cache_contexts/prune_config.py,sha256=OdMEs_nBO3SDq8vC4JcF4LyINws2QWxFZOkh9N53ScQ,3204
+cache_dit/caching/cache_contexts/prune_context.py,sha256=HlPARnwrRQmkgdcsAYSm7_nhe8X7-dQYrIxbyVW64x0,6341
+cache_dit/caching/cache_contexts/prune_manager.py,sha256=KOn6h2IJmM5GUMUzCNjM6xgmaABzQ_zIV-BuAxL0bBM,6111
+cache_dit/caching/cache_contexts/calibrators/__init__.py,sha256=Kgyy93ftAaJp-GJ1XtJqQ9AaN8rlaMCHcFLwDeKZhtI,6251
+cache_dit/caching/cache_contexts/calibrators/base.py,sha256=mn6ZBkChGpGwN5csrHTUGMoX6BBPvqHXSLbIExiW-EU,748
+cache_dit/caching/cache_contexts/calibrators/foca.py,sha256=dh0LsV_y9RkDxmcL_VaUYTEylkzSuqlBHKUKnSxTGZU,885
+cache_dit/caching/cache_contexts/calibrators/taylorseer.py,sha256=lN2y_baGCG-aRed_xTVjybrOyCH4YivKpQbxQUMHT24,6003
+cache_dit/caching/patch_functors/__init__.py,sha256=mP81SWtlqUGjLMLXIpAN7_R0GW_RZBVv_Rr01IB0v7k,610
+cache_dit/caching/patch_functors/functor_base.py,sha256=Ahk0fTfrHgNdEl-9JSkACvfyyv9G-Ei5OSz7XBIlX5o,357
+cache_dit/caching/patch_functors/functor_chroma.py,sha256=BNhFkSQe8PCZ6aK68xQYcosIeDByxxRF-e5ZE7NZXHg,13424
+cache_dit/caching/patch_functors/functor_dit.py,sha256=wbo8cQZbgtLFAHHFZb5bXMQ572jr9UK7VmVl-MsG_wM,3898
+cache_dit/caching/patch_functors/functor_flux.py,sha256=xyPY2FJ1wy1UJfgX_MVBCJvmH2hydNjHbeyRFr2N39Y,9553
+cache_dit/caching/patch_functors/functor_hidream.py,sha256=OO3_cVTGbG1VKnCHU-nu9k6VMpQKa-oQWVghT64FuyM,15303
+cache_dit/caching/patch_functors/functor_hunyuan_dit.py,sha256=mrSpxbdUY7RY_EEH4X22RGHI00BeTUFOL_-G33vTyBA,6395
+cache_dit/caching/patch_functors/functor_qwen_image_controlnet.py,sha256=FxPcD_fCdJ-9AFn2EMbI32oi61r6Y-3A4W_U-TbCTLA,10513
+cache_dit/compile/__init__.py,sha256=FcTVzCeyypl-mxlc59_ehHL3lBNiDAFsXuRoJ-5Cfi0,56
+cache_dit/compile/utils.py,sha256=nN2OIrSdwRR5zGxJinKDqb07pXpvTNTF3g_OgLkeeBU,3858
+cache_dit/kernels/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cache_dit/kernels/triton_taylorseer.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cache_dit/metrics/__init__.py,sha256=Y_JrBr9XE6NKXwyXc7d_-PaX9c_rk5FKms-IYgCyHmY,936
+cache_dit/metrics/clip_score.py,sha256=ERNCFQFJKzJdbIX9OAg-1LiSPuXUVHLOFxbf2gcENpc,3938
+cache_dit/metrics/config.py,sha256=ieOgD9ayz722RjVzk24bSIqS2D6o7TZjGk8KeXV-OLQ,551
+cache_dit/metrics/fid.py,sha256=ZM_FM0XERtpnkMUfphmw2aOdljrh1uba-pnYItu0q6M,18219
+cache_dit/metrics/image_reward.py,sha256=N8HalJo1T1js0dsNb2V1KRv4kIdcm3nhx7iOXJuqcns,5421
+cache_dit/metrics/inception.py,sha256=pBVe2X6ylLPIXTG4-GWDM9DWnCviMJbJ45R3ulhktR0,12759
+cache_dit/metrics/lpips.py,sha256=hrHrmdM-f2B4TKDs0xLqJO5JFaYcCjq2qNIR8oCrVkc,811
+cache_dit/metrics/metrics.py,sha256=AZbQyoavE-djvyRUZ_EfCIrWSQbiWQFo7n2dhn7XptE,40466
+cache_dit/parallelism/__init__.py,sha256=dheBG5_TZCuwctviMslpAEgB-B3N8F816bE51qsw_fU,210
+cache_dit/parallelism/parallel_backend.py,sha256=XFmv8GmAKfadJQomUbAgox9aJQxmNxdlPhVE6nJBn0s,939
+cache_dit/parallelism/parallel_config.py,sha256=kgpysG1lKCJ6Wd3llQFSmiY_Hdorv2wTfm2Xcj6qec0,3738
+cache_dit/parallelism/parallel_interface.py,sha256=2OtZiXmpQbSchzH_MKUwMr1iDB3gj1sklQJGA_Kc04I,2664
+cache_dit/parallelism/backends/native_diffusers/__init__.py,sha256=rpYu52_ATmsBpztWSzpwpYvhsQ0ABJe-KyP4UyawgZ8,237
+cache_dit/parallelism/backends/native_diffusers/parallel_difffusers.py,sha256=CY53v3lc36o8HXdc42xc5c0wsKdpDiswNDM3kLAuW-0,1607
+cache_dit/parallelism/backends/native_diffusers/utils.py,sha256=VAqtv9b8PTvcoYzD_CbvtRgdg9_VbtBug_5L38PADl0,266
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/__init__.py,sha256=oOCVeisY7K8JPw5JaGmqhoZZslFDAsYC5oNgbTmt1vU,6228
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_chroma.py,sha256=r1LbdyI26sY1Ig5sX24qtUNdk5NfX-DLP0Po2_cCR78,3899
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_cogvideox.py,sha256=QKS0xN62N7mTlJIutuOoiuYW9vkjmwrjZM8quPbaDKw,7812
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_cogview.py,sha256=fsifJF1Nlh96YVVYk9qa4q7xvno3gqWncpZkrMmCQZY,12403
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_cosisid.py,sha256=ry0zmbbVFiu_sBXXGp4XCO9rzZTxJ9dORFVqXIfLUR8,5248
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_dit.py,sha256=jrrSciQUYP4T4bc7BuXVt6ha1VJC-L3jrcNywVHIRng,3346
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_flux.py,sha256=yaGCqywRGjd_UL8FtwnEbX3NmM7z2XPhMeo3TNfI7mA,3690
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_hunyuan.py,sha256=2VcObE9F25tF9NJNTudcQX8nYediF3CvF2GYE85Lm60,28628
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_ltxvideo.py,sha256=tzIx3MOMGuO2hVIhaM_z6IXRdYABX4sjfqd5slhRCVc,10470
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_nunchaku.py,sha256=VvE4tzKqLU5h4CxyFT1NG2xamcDy11iQQS_NsbYrEbA,16712
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_pixart.py,sha256=vftDDzQ62PXCl3cji8uFV9OpQSfX0aP7X0QlNW1mJzM,10872
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_qwen_image.py,sha256=XbNE40TMC_RptsXJXIGEWTPTlccQPG5z0oLGWhQKlYE,4314
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_registers.py,sha256=W3Uh71YMxwqUqJZp8wrDN_E1Xcxi5UNRKJefPjFm15g,2713
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_plan_wan.py,sha256=DZlt3r04KfCpFy43JXO2OTDNqsgHW282mgBh5ViL6yc,3977
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/cp_planners.py,sha256=_dTHUdlhVunZOwidDN5zqHtrwLxv6ykZW9TApMTcaqA,5206
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/attention/__init__.py,sha256=6jIdEQGxC3I97MjLmxkXr3bp6QhzGQ7nId-PzA9ffbg,246
+cache_dit/parallelism/backends/native_diffusers/context_parallelism/attention/_attention_dispatch.py,sha256=a6_9PDfBfhQo3fkRpB7y6vDK1vHAmkHTrX3Z0WZ29og,10411
+cache_dit/parallelism/backends/native_pytorch/__init__.py,sha256=EAzdGEP47-y9w6uyPisvDjCF3TE03ZCdhalAY89p-ZA,226
+cache_dit/parallelism/backends/native_pytorch/parallel_torch.py,sha256=1A-Gdv6JRueZNpwJdrVxRT7TBxwo2Bn-CwMbnlhJx0M,2047
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/__init__.py,sha256=M-nEf6a22UeoIjZEhIajpUpGSQzWiNn_zmWiBNU70Fs,1662
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/tp_plan_flux.py,sha256=yiZeyFigd3kpHLMGQ6tt3iK9CU6gHaLm3MiVZWEb4_0,6549
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/tp_plan_kandinsky5.py,sha256=MwNzjj8nyxOzqdmyhAaeslFM5iVHBY0I2Jd9W4TjTjs,2702
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/tp_plan_qwen_image.py,sha256=AcRCfML0bwGeVY0gnVfMQtW9GAWc5wO4ZhxWGNDCkKc,2644
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/tp_plan_registers.py,sha256=ERgUE6RnweM7BGp9aO4jywAyMrS8KHrL1mTL7XReMkw,2099
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/tp_plan_wan.py,sha256=j8w5gZn3MhSsacxOVguss85GSI6bdLL61FMySDZKwJc,5191
+cache_dit/parallelism/backends/native_pytorch/tensor_parallelism/tp_planners.py,sha256=GbhB7-21DiNF786oGw-AeGBUISkjzEgp82Mztjkl6J0,571
+cache_dit/quantize/__init__.py,sha256=rUu0V9VRjOgwXuIUHHAI-osivNjAdUsi-jpkDbFp6Gk,278
+cache_dit/quantize/quantize_backend.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cache_dit/quantize/quantize_config.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cache_dit/quantize/quantize_interface.py,sha256=dKLrLF-FDdRLQq-3CnaRzkAh70P4oObJE3-qWF7goM0,882
+cache_dit/quantize/backends/__init__.py,sha256=SL9EupOwBRzRcHZBI1ABqdHjCS9vEpFZXjA9R5ikTk8,33
+cache_dit/quantize/backends/bitsandbytes/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+cache_dit/quantize/backends/torchao/__init__.py,sha256=NK1widhJeKSx8ICCcyYZAWGUpGc2uKF5O7pLIcSCUDI,37
+cache_dit/quantize/backends/torchao/quantize_ao.py,sha256=-a99fEp5xjEXpIgD4Hu1lBVk7bBl2XN1tsWhGlK2cN4,6718
+cache_dit-1.0.14.dist-info/licenses/LICENSE,sha256=6LIQc3N_o1SeNXCONTZUJPO4bFIYccq-E4Qp8Dti5CQ,11438
+cache_dit-1.0.14.dist-info/METADATA,sha256=Rhfwk2kYNDQuXKS9VnE_rgClkf1qwmP62ERGK1HqwkI,31359
+cache_dit-1.0.14.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+cache_dit-1.0.14.dist-info/entry_points.txt,sha256=FX2gysXaZx6NeK1iCLMcIdP8Q4_qikkIHtEmi3oWn8o,65
+cache_dit-1.0.14.dist-info/top_level.txt,sha256=ZJDydonLEhujzz0FOkVbO-BqfzO9d_VqRHmZU-3MOZo,10
+cache_dit-1.0.14.dist-info/RECORD,,