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 17, 05:30 UTC
Miner fix of type hint (#2340)
ljss committed
Jan 4, 2024 at 13:27 UTC
aee8ef661a227bfeb0c886b2b3892b0fc56bbd47
1 file changed
+1
-1
vllm/engine/ray_utils.py
+1
-1
@@ -55,7 +55,7 @@ def initialize_cluster(
55
parallel_config: ParallelConfig,
56
engine_use_ray: bool = False,
57
ray_address: Optional[str] = None,
58
-) -> Tuple[str, Optional["PlacementGroup"]]:
58
+) -> Optional["PlacementGroup"]:
59
"""Initialize the distributed cluster probably with Ray.
60
61
Args: