Update README.md

README.md

@@ -23,6 +23,10 @@ To run the demo, use this notebook on Colab:
 
 https://colab.research.google.com/github/wdcqc/WaveFunctionDiffusion/blob/remaster/colab/WaveFunctionDiffusion_Demo.ipynb
 
+Alternatively run it on Huggingface Spaces: (but it is slow, so it's recommended to run on Colab)
+
+https://huggingface.co/spaces/wdcqc/wfd
+
 In addition to Dreambooth, a custom VAE model (`AutoencoderTile`) is trained to encode and decode the latents to/from tileset probabilities ("waves") and then generated as Starcraft maps.
 
 A WFC Guidance, inspired by the Wave Function Collapse algorithm, is also added to the pipeline. For more information about guidance please see this page: [Fine-Tuning, Guidance and Conditioning](https://github.com/huggingface/diffusion-models-class/tree/main/unit2)