A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 22:15 UTC
[Misc]Code Cleanup (#13859)
Signed-off-by: noemotiovon <noemotiovon@gmail.com> Co-authored-by: noemotiovon <noemotiovon@gmail.com>
Chenguang Li committed
Feb 26, 2025 at 10:44 UTC
e1fe7591f222753e101487e8032c2e08a5f07d61
1 file changed
-1
vllm/executor/ray_distributed_executor.py
-1
@@ -95,7 +95,6 @@ class RayDistributedExecutor(DistributedExecutorBase):
95
self.use_v1 = envs.VLLM_USE_V1
96
97
self.pp_locks: Optional[List[asyncio.Lock]] = None
98
- self.use_ray_spmd_worker = envs.VLLM_USE_RAY_SPMD_WORKER
98
if not self.use_ray_compiled_dag:
99
self.driver_exec_method = make_async(
100
self.driver_worker.execute_method)