[V1][PP] Enable true PP with Ray executor (#13472)

Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
2025-02-18 09:15:32 -08:00
parent a02c86b4dd
commit 4fb8142a0e
1 changed files with 1 additions and 1 deletions
--- 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,