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, 18:15 UTC
[ci][test] fix pp test failure (#7945)
youkaichao committed
Aug 28, 2024 at 01:27 UTC
f52a43a8b90f8c4d5ba63003cc9ae75701ad48d9
1 file changed
+1
-1
vllm/executor/multiproc_gpu_executor.py
+1
-1
@@ -126,7 +126,7 @@ class MultiprocessingGPUExecutor(DistributedGPUExecutor):
126
max_parallel_loading_workers)
127
128
def _check_executor_parameters(self):
129
- world_size = self.parallel_config.tensor_parallel_size
129
+ world_size = self.parallel_config.world_size
130
tensor_parallel_size = self.parallel_config.tensor_parallel_size
131
132
# Set CUDA_VISIBLE_DEVICES for the driver, inherited by workers