Add HuggingFace Stable Diffusion pipeline demo

Signed-off-by: Kevin Chen <kevinch@nvidia.com>

Author: kevinch-nv

demo/experimental/HuggingFace-Diffusers/README.md

@@ -0,0 +1,36 @@
+# Introduction
+
+This demo notebook showcases the acceleration of Stable Diffusion pipeline using TensorRT through HuggingFace pipelines.
+
+# Setup
+
+### Clone the TensorRT OSS repository
+
+```bash
+git clone git@github.com:NVIDIA/TensorRT.git -b release/8.6 --single-branch
+cd TensorRT/demo/experimental/HuggingFace-Diffusers
+```
+
+### Launch TensorRT NGC container
+
+Install nvidia-docker using [these intructions](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#docker). Launch the docker container with the following command:
+
+```bash
+docker run --rm -it --gpus all -p 8888:8888 -v $PWD:/workspace nvcr.io/nvidia/tensorrt:23.04-py3 /bin/bash
+```
+
+### Run Jupyter Notebook
+
+Install `jupyter` with:
+
+```bash
+pip install jupyter
+```
+
+Launch the notebook within the container with:
+
+```bash
+jupyter notebook --ip 0.0.0.0 TensorRT-diffusers-txt2img.ipynb --allow-root --no-browser
+```
+
+Follow the console output for the link to run the notebook on your host machine.