Code
Hub
Workspaces
Connect
Indexed graphs
Engine
MCP
copy
hub
/
github.com/allenai/open-instruct
/ sleep
Method
sleep
open_instruct/vllm_utils.py:788–789 ·
view source on GitHub ↗
(self)
Source
from the content-addressed store, hash-verified
786
return
future.result()
787
788
def
sleep(self) -> None:
789
return
self._run_async(self.llm_engine.sleep(level=0, mode=
"keep"
))
790
791
def
wake_up(self) -> None:
792
return
self._run_async(self.llm_engine.wake_up(tags=[
"scheduling"
]))
Callers
15
test_streaming_dataloader_with_rollout_saving
Method · 0.80
wrapper
Function · 0.80
async_call
Method · 0.80
setUpClass
Method · 0.80
setUpClass
Method · 0.80
wait_for_grpo_fast_minimum_cluster_resources
Function · 0.80
health_check_fn
Function · 0.80
_prefetch_worker
Function · 0.80
_init_engine_and_server
Method · 0.80
update_weights
Method · 0.80
slow_task
Method · 0.80
_data_preparation_loop
Method · 0.80
Calls
1
_run_async
Method · 0.95
Tested by
10
test_streaming_dataloader_with_rollout_saving
Method · 0.64
setUpClass
Method · 0.64
setUpClass
Method · 0.64
slow_task
Method · 0.64
test_timer_context_manager
Method · 0.64
slow_function
Method · 0.64
test_timer_noop
Method · 0.64
silent_function
Method · 0.64
start
Method · 0.64
acompletion
Method · 0.64