[v1] Refactor KVCacheConfig (#14079)

Signed-off-by: Chen Zhang <zhangch99@outlook.com>
2025-03-21 19:56:27 +08:00
parent 61e8c18350
commit 93a00d7dde
10 changed files with 318 additions and 110 deletions
--- a/vllm/v1/worker/gpu_worker.py
+++ b/vllm/v1/worker/gpu_worker.py
@@ -185,7 +185,7 @@ class Worker(WorkerBase):

        return int(available_kv_cache_memory)

-    def get_kv_cache_spec(self) -> KVCacheSpec:
+    def get_kv_cache_spec(self) -> dict[str, KVCacheSpec]:
        return self.model_runner.get_kv_cache_spec()

    def initialize_from_config(self, kv_cache_config: KVCacheConfig) -> None: