Update README.md

README.md

@@ -5,7 +5,9 @@ inference: false
 tags:
 - deepsparse
 ---
-This is a [SparseML](https://github.com/neuralmagic/sparseml) quantized version of https://huggingface.co/laion/CLIP-ViT-B-32-256x256-DataComp-s34B-b86K that is ready to use with [DeepSparse](https://github.com/neuralmagic/deepsparse). It achieves 71.1% zero-shot top-1 accuracy on ImageNet and 95.6% zero-shot top-1 accuracy on Imagenette.
+This is a [SparseML](https://github.com/neuralmagic/sparseml) quantized version of https://huggingface.co/laion/CLIP-ViT-B-32-256x256-DataComp-s34B-b86K that is ready to use with [DeepSparse](https://github.com/neuralmagic/deepsparse). 
+It achieves **71.1%** zero-shot top-1 accuracy on ImageNet and **95.6%** zero-shot top-1 accuracy on Imagenette. 
+For comparison the dense version (the original model) achieves **72.8%** on ImageNet and **95.7%** on Imagenette.
 
 Notebook for basic usage: [![Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1ZvU9ZSHJKSeJyH5bgxo_A-GSVIUcSt2E?usp=sharing)
 Notebook for Imagenette evaluation: [![Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1-Duq0YNtjzOnmuXCYo-5DDiOzeCItXpN?usp=sharing)