From 4fb8142a0e47c83c5f98fcf76f1168d6f6057ff3 Mon Sep 17 00:00:00 2001 From: Woosuk Kwon Date: Tue, 18 Feb 2025 09:15:32 -0800 Subject: [PATCH] [V1][PP] Enable true PP with Ray executor (#13472) Signed-off-by: Woosuk Kwon --- vllm/v1/executor/ray_distributed_executor.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/v1/executor/ray_distributed_executor.py b/vllm/v1/executor/ray_distributed_executor.py index 53548610adf6..320ebfd37ae3 100644 --- a/vllm/v1/executor/ray_distributed_executor.py +++ b/vllm/v1/executor/ray_distributed_executor.py @@ -32,7 +32,7 @@ class RayDistributedExecutor(RayDistributedExecutorV0, Executor): """Ray distributed executor supports pipeline parallelism, meaning that it allows PP size batches to be executed concurrently. """ - return 1 #self.vllm_config.parallel_config.pipeline_parallel_size + return self.parallel_config.pipeline_parallel_size def execute_model( self,