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 16, 19:30 UTC
[Bugfix][CI] Import ray under guard (#7486)
Woosuk Kwon committed
Aug 13, 2024 at 17:12 UTC
59edd0f1340a85d585a477fc687a16cfc5cf5276
1 file changed
+1
-1
vllm/distributed/device_communicators/tpu_communicator.py
+1
-1
@@ -1,4 +1,3 @@
1
-import ray
1
import torch
2
import torch.distributed as dist
3
from torch.distributed import ProcessGroup
@@ -6,6 +5,7 @@ from torch.distributed import ProcessGroup
5
from vllm.platforms import current_platform
6
7
if current_platform.is_tpu():
8
+ import ray
9
import torch_xla.core.xla_model as xm
10
import torch_xla.runtime as xr
11
from torch_xla._internal import pjrt