Create README.md

README.md

@@ -0,0 +1,41 @@
+---
+license: apache-2.0
+---
+<h2 align="center">lyraDiff: An Out-of-box Acceleration Engine for Diffusion and DiT Models</h2>
+<div align="center">   
+Sa Xiao<sup>*</sup>, Yibo Lu<sup>*</sup>,  Kangjian Wu<sup>*</sup>, Bin Wu<sup>†</sup>, Haoxiong Su, Mian Peng, Qiwen Mao, Wenjiang Zhou</br>(*co-first author), (†Corresponding Author, benbinwu@tencent.com)</br>
+Lyra Lab, Tencent Music Entertainment</br>     
+<p>[<a href="https://github.com/TMElyralab/lyraDiff">github</a>] [<a href="https://huggingface.co/TMElyralab/lyraDiff">huggingface</a>] </p>
+</div>    
+
+## Introduction
+
+🌈`lyraDiff` is currently the **Fastest Diffusion Acceleration Engine** that **doesn't need recompilation with dynamic input shapes**.
+
+The core features include:
+- 🚀 **State-of-the-art Inference Speed**: `lyraDiff` utilizes multiple techniques to achieve up to 2x speedup of the model inference, including **Quantization**, **Fused GEMM Kernels**, **Flash Attention**, and **NHWC & Fused GroupNorm**.
+- 🔥 **Memory Efficiency**: `lyraDiff` utilizes buffer-based DRAM reuse strategy and multiple types of quantizations (FP8/INT8/INT4) to save **10-40%** of DRAM usage.
+- 🔥 **Extensive Model Support**: `lyraDiff` supports a wide range of Generative/SR models such as **SD1.5, SDXL, FLUX, S3Diff, SUPIR, etc.**, and those most commonly used plugins such as **LoRA, ControlNet and Ip-Adapter**.
+- 🔥 **Zero Compilation Deployment**: Unlike **TensorRT** or **AITemplate**, which takes minutes to compile, `lyraDiff` eliminates runtime recompilation overhead even with model inputs of dynamic shapes.
+- 🔥 **Image Gen Consistency**: The outputs of `lyraDiff` are aligned with the ones of [HF diffusers](https://github.com/huggingface/diffusers) at the pixel level, even under LoRA switch in quantization mode.
+- 🚀 **Fast Plugin Hot-swap**: `lyraDiff` provides **Super Fast Model Hot-swap for ControlNet and LoRA** which can hugely benefit a real-time image gen service.
+
+
+![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6461b412846a6c8c8305319d/IAcfgrmYOcRS9vcNqvkzH.jpeg)
+
+`lyraDiff-IP-Adapters` converted from the standard [IP-Adapter](https://huggingface.co/h94/IP-Adapter) weights using this [script](https://github.com/TMElyralab/lyraDiff/blob/main/lyradiff/convert_model_scripts/convert_ipadapter.py) to be compatiable with [lyraDiff](https://github.com/TMElyralab/lyraDiff)
+
+It contains both SD1.5 and SDXL version of converted IP-Adapter
+
+## Usage
+
+We provide a reference implementation of lyraDiff version of SD1.5/SDXL, as well as sampling code, in a dedicated [github repository](https://github.com/TMElyralab/lyraDiff).
+
+## Citation
+``` bibtex
+@Misc{lyraDiff_2025,
+  author =       {Kangjian Wu, Zhengtao Wang, Yibo Lu, Haoxiong Su, Sa Xiao, Qiwen Mao, Mian Peng, Bin Wu, Wenjiang Zhou},
+  title =        {lyraDiff: Accelerating Diffusion Models with best flexibility},
+  howpublished = {\url{https://github.com/TMElyralab/lyraDiff}},
+  year =         {2025}
+}