rename multi level kv cache

Author: liujiacheng

lightllm/server/router/model_infer/mode_backend/base_backend.py

@@ -31,7 +31,7 @@
 from lightllm.server.router.shm_reqs_io_buffer import ShmReqsIOBuffer
 from lightllm.server.router.model_infer.mode_backend.overlap_events import OverlapEventManager, OverlapEventPack
 from lightllm.models.deepseek_mtp.model import Deepseek3MTPModel
-from .multi_level_cache import MultiLevelCacheModule
+from .multi_level_kv_cache import MultiLevelKvCacheModule
 
 
 class ModeBackend:
@@ -201,7 +201,7 @@ def init_model(self, kvargs):
         self.infer_loop_thread1.start()
 
         if self.args.enable_cpu_cache:
-            self.multi_level_cache_module = MultiLevelCacheModule(self)
+            self.multi_level_cache_module = MultiLevelKvCacheModule(self)
         return
 
     def init_custom(self):

…_infer/mode_backend/multi_level_cache.py …fer/mode_backend/multi_level_kv_cache.pylightllm/server/router/model_infer/mode_backend/multi_level_cache.py renamed to lightllm/server/router/model_infer/mode_backend/multi_level_kv_cache.py lightllm/server/router/model_infer/mode_backend/multi_level_cache.py renamed to lightllm/server/router/model_infer/mode_backend/multi_level_kv_cache.py

@@ -12,7 +12,7 @@
 from lightllm.server.router.model_infer.infer_batch import g_infer_context
 
 
-class MultiLevelCacheModule(object):
+class MultiLevelKvCacheModule(object):
     def __init__(self, backend):
         self.args = get_env_start_args()
         from .base_backend import ModeBackend