Update README.md

README.md

@@ -27,22 +27,20 @@ metrics:
 
 ## Setup
 
-You can run the smashed model on a A100 GPU by:
-1. Installing and importing the `pruna-engine` (version 0.2.7) package. Use `pip install pruna --extra-index-url https://pypi.nvidia.com --extra-index-url https://pypi.ngc.nvidia.com` for installation. See [Pypi](https://pypi.org/project/pruna-engine/) for detailed on the package.
-2. Downloading the model files at `model_path`. This can be done using huggingface with this repository name or with manual downloading.
-3. Loading the model
-4. Running the model. 
-
-The model is smashed for 768x768 image size. Contact us if you want a version with dynamic size support.
-
+You can run the smashed model by:
+1. Installing and importing the `pruna-engine` (version 0.2.9) package. Use `pip install pruna --extra-index-url https://pypi.nvidia.com --extra-index-url https://pypi.ngc.nvidia.com` for installation. See [Pypi](https://pypi.org/project/pruna-engine/) for details on the package.
+2. Downloading the model files. This can be done using the Hugging Face CLI with the following commands:
+    ```bash
+    mkdir stabilityai-stable-diffusion-2-1-base-smashed
+    huggingface-cli download PrunaAI/stabilityai-stable-diffusion-2-1-base-smashed --local-dir stabilityai-stable-diffusion-2-1-base-smashed --local-dir-use-symlinks False
+    ```
+   Alternatively, you can download them manually.
+3. Loading the model.
+4. Running the model.
 You can achieve this by running the following code:
-
 ```python
-from transformers.utils.hub import cached_file
 from pruna_engine.PrunaModel import PrunaModel  # Step (1): install and import `pruna-engine` package.
-
-...
-model_path = cached_file("PrunaAI/stabilityai-stable-diffusion-2-1-base-smashed", "model")  # Step (2): download the model files at `model_path`.
+model_path = "stabilityai-stable-diffusion-2-1-base-smashed/model" # Step (2): specify the downloaded model path.
 smashed_model = PrunaModel.load_model(model_path)  # Step (3): load the model.
 y = smashed_model(prompt="a silly prune with a face in high definition", image_height=768, image_width=768)[0]  # Step (4): run the model.
 ```