Update README.md

README.md

@@ -20,7 +20,7 @@ These are RWKV-4-Pile 1.5/3/7/14B models finetuned on Alpaca, CodeAlpaca, Guanac
 
 Gradio Demo: https://huggingface.co/spaces/BlinkDL/Raven-RWKV-7B
 
-Use https://github.com/BlinkDL/ChatRWKV (fast CUDA) or https://github.com/saharNooby/rwkv.cpp (fast CPU INT4) to run them.
+Use https://github.com/BlinkDL/ChatRWKV (fast CUDA) or https://github.com/saharNooby/rwkv.cpp (fast CPU, use INT8 for best results) to run them.
 
 See https://github.com/BlinkDL/RWKV-LM for details on the RWKV Language Model (100% RNN).