Update README.md

README.md

@@ -19,13 +19,13 @@ This is a Stable Diffusion model fine-tuned on Starcraft terrain images on the S
 
 It was trained on 32x32 terrain images from 265 melee maps including original Blizzard maps and those downloaded from Battle.net, scmscx.com and broodwarmaps.net.
 
-To run the demo, use this notebook on Colab:
+Run it on Huggingface Spaces:
 
-https://colab.research.google.com/github/wdcqc/WaveFunctionDiffusion/blob/remaster/colab/WaveFunctionDiffusion_Demo.ipynb
+https://huggingface.co/spaces/wdcqc/wfd
 
-Alternatively run it on Huggingface Spaces: (it is slow, recommended to run on Colab)
+Or use this notebook on Colab:
 
-https://huggingface.co/spaces/wdcqc/wfd
+https://colab.research.google.com/github/wdcqc/WaveFunctionDiffusion/blob/remaster/colab/WaveFunctionDiffusion_Demo.ipynb
 
 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.