init qwen-image-fast

Author: DefTruth

README.md

@@ -1,7 +1,7 @@
 <div align="center">
   <p align="center">
   <h1>⚡️Qwen-Image-Fast</h1> 
-  🔥Qwen-Image <b>4.8x</b> speedup w/ <b>Hybrid Acceleration</b> for <b>low</b> VRAM GPUs (< 48GiB)
+  🔥Qwen-Image <b>4.8x🎉</b> speedup with <b>Hybrid Acceleration</b> for <b>low</b> VRAM GPUs (< 48GiB)
 </div>
 
 ## ⚙️Installation
@@ -32,13 +32,10 @@ torchrun --nproc_per_node=2 qwen_image_fast.py \
          --compile
 ```
 
-### 🤖Baseline w/o Acceleration: ~120s  
-
-<img width="1024" height="1024" alt="Image" src="https://github.com/user-attachments/assets/d020be51-1950-4d4f-b390-5ebe8a835430" />
-
-### 🎉Hybrid Acceleration: 25s, ~4.8x speedup
-
-<img width="1024" height="1024" alt="Image" src="https://github.com/user-attachments/assets/8de94d39-88c4-4c00-b323-b2308836ad04" />
+|🤖Baseline w/o Acceleration|🎉w/ Hybrid Acceleration|
+|:---:|:---:|
+|~120s, 60+ GiB per GPU|~25s, ~4.8x speedup, 36 GiB per GPU| 
+|![](./assets/qwen-image-fast.C0_L0_Q0_NONE.png)|![](./assets/qwen-image-fast.C1_L0_Q1_DBCache_F1B0_W8I1M0MC2_R0.12_T1O1_Ulysses2_S24.png)|
 
 ## ©️Acknowledgements