Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions vllm/entrypoints/llm.py
Original file line number Diff line number Diff line change
Expand Up @@ -1157,6 +1157,9 @@ def sleep(self, level: int = 1):
self.llm_engine.sleep(level=level)

def wake_up(self):
"""
Wake up the engine from sleep mode. See the :meth:`sleep` method
for more details."""
self.llm_engine.wake_up()

# LEGACY
Expand Down
9 changes: 9 additions & 0 deletions vllm/executor/executor_base.py
Original file line number Diff line number Diff line change
Expand Up @@ -47,6 +47,7 @@ def __init__(
self.prompt_adapter_config = vllm_config.prompt_adapter_config
self.observability_config = vllm_config.observability_config
self._init_executor()
self.is_sleeping = False

@abstractmethod
def _init_executor(self) -> None:
Expand Down Expand Up @@ -194,10 +195,18 @@ def stop_profile(self) -> None:
self.collective_rpc("stop_profile")

def sleep(self, level: int = 1):
if self.is_sleeping:
logger.warning("Executor is already sleeping.")
return
self.collective_rpc("sleep", kwargs=dict(level=level))
self.is_sleeping = True

def wake_up(self):
if not self.is_sleeping:
logger.warning("Executor is not sleeping.")
return
self.collective_rpc("wake_up")
self.is_sleeping = False

def save_sharded_state(
self,
Expand Down
Loading