update doc

README.md

@@ -48,7 +48,7 @@ Below is average latency of generating an image of size 512x512 using NVIDIA A10
 
 
 Static means the engine is built for the given batch size and image size combination, and CUDA graph is used to speed up.
-
+For PyTorch 2.1, the UNet use channel last (NHWC) format, and compile the UNet with mode `reduce-overhead`. See [benchmark script](https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/python/tools/transformers/models/stable_diffusion/benchmark_controlnet.py) for detail.
 
 #### Latency for SDXL-Turbo with Canny Control Net
 
@@ -56,10 +56,10 @@ Below is average latency of generating an image of size 512x512 with canny contr
 
 | Engine      | Batch Size | Steps | PyTorch 2.1     | ONNX Runtime CUDA |
 |-------------|------------|------ | ----------------|-------------------|
-| Static      | 1          |   1   | 160.0 ms        |  49.3 ms          |
-| Static      | 4          |   1   | 314.9 ms        | 135.3 ms          |
-| Static      | 1          |   4   | 251.9 ms        | 123.3 ms          |
-| Static      | 4          |   4   | 514.2 ms        | 303.3 ms          |
+| Static      | 1          |   1   | 160.0 ms        |  55.3 ms          |
+| Static      | 4          |   1   | 314.9 ms        | 144.4 ms          |
+| Static      | 1          |   4   | 251.9 ms        | 134.9 ms          |
+| Static      | 4          |   4   | 514.2 ms        | 332.6 ms          |
 
 
 ## Usage Example