[V1] Allow sliding window + prefix caching (vllm-project#13069)

Signed-off-by: Woosuk Kwon <[email protected]>

Author: WoosukKwon

vllm/config.py

@@ -1146,7 +1146,7 @@ def _verify_prefix_caching(self) -> None:
         if not self.enable_prefix_caching:
             return
 
-        if self.sliding_window is not None:
+        if self.sliding_window is not None and not envs.VLLM_USE_V1:
             raise NotImplementedError(
                 "Prefix caching is not supported with sliding window. "
                 "Run with --disable-sliding-window to use prefix caching.")