Update README.md

README.md

@@ -15,7 +15,7 @@ The inference speed of lyraBELLE has achieved **3.3x+** acceleration upon the or
 Among its main features are:
 
 - weights: the original BELLE-7B-2M weights released by BelleGroup.
-- device: Nvidia Ampere architechture or newer (e.g A100)
+- device: Nvidia Ampere architechture or newer (e.g., A100)
 
 Note that:
 **Some interface/code were set for future uses(see demo below).**