diff --git a/README.md b/README.md
index 97cbf77..96849f9 100644
--- a/README.md
+++ b/README.md
@@ -51,6 +51,8 @@ Use https://github.com/BlinkDL/RWKV-LM/tree/main/RWKV-v4 for the latest RWKV-4 m
 
 ### Inference
 
+**Colab for RWKV-4 Pile 1.5B**: https://colab.research.google.com/drive/1F7tZoPZaWJf1fsCmZ5tjw6sYHiFOYVWM
+
 **Run RWKV-4 Pile models:** Download models from https://huggingface.co/BlinkDL. Set TOKEN_MODE = 'pile' in run.py and run it. It's fast even on CPU (the default mode).
 
 Run RWKV-4 Pile models in your browser (and onnx version): see this issue https://github.com/BlinkDL/RWKV-LM/issues/7