Update README.md

README.md

@@ -46,10 +46,54 @@ One single flow of Versatile Diffusion contains a VAE, a diffuser, and a context
       }
 ```
 
+# Usage
+
 You can use the model both with the [🧨Diffusers library](https://github.com/huggingface/diffusers) and the [SHI-Labs Versatile Diffusion codebase](https://github.com/SHI-Labs/Versatile-Diffusion).
 
-### Diffusers
-#### Text to Image
+
+
+## 🧨 Diffusers
+
+🧨 Diffusers let's you both use a unified and more memory-efficient, task-specific pipelines. 
+
+## VersatileDiffusionPipeline
+
+To use Versatile Diffusion for all tasks, it is recommend to use the [`VersatileDiffusionPipeline`](https://huggingface.co/docs/diffusers/main/en/api/pipelines/versatile_diffusion#diffusers.VersatileDiffusionPipeline)
+
+```py
+from diffusers import VersatileDiffusionPipeline
+import torch
+import requests
+from io import BytesIO
+from PIL import Image
+
+# let's download an initial image
+url = "https://huggingface.co/datasets/diffusers/images/resolve/main/benz.jpg"
+
+response = requests.get(url)
+image = Image.open(BytesIO(response.content)).convert("RGB")
+
+pipe = VersatileDiffusionPipeline.from_pretrained(
+    "shi-labs/versatile-diffusion", torch_dtype=torch.float16
+)
+pipe = pipe.to("cuda")
+
+generator = torch.Generator(device="cuda").manual_seed(0)
+image = pipe.image_variation(image, generator=generator).images[0]
+image.save("./car_variation.png")
+
+#similarly with 
+# pipe.text_to_image(...)
+# pipe.dual_guided(...)
+```
+
+### Task Specific
+
+The task specific pipelines load only the weights that are needed onto GPU. 
+You can find all task specific pipelines [here](https://huggingface.co/docs/diffusers/main/en/api/pipelines/versatile_diffusion#versatilediffusion).
+
+You can use them as follows:
+### Text to Image
 ```py
 from diffusers import VersatileDiffusionTextToImagePipeline
 import torch