[Core] Support reset_prefix_cache (#12284)

2025-01-22 10:52:27 -08:00
parent 96f6a7596f
commit 7206ce4ce1
27 changed files with 300 additions and 21 deletions
--- a/vllm/v1/engine/llm_engine.py
+++ b/vllm/v1/engine/llm_engine.py
@@ -162,6 +162,9 @@ class LLMEngine:
    def stop_profile(self):
        self.engine_core.profile(False)

+    def reset_prefix_cache(self):
+        self.engine_core.reset_prefix_cache()
+
    def get_tokenizer_group(
        self,
        group_type: Type[_G] = BaseTokenizerGroup,