From a1d82466ea2b174768e20874d9eb1a689a0e675d Mon Sep 17 00:00:00 2001 From: Ifta khairul Alam Adil <25082512+ikaadil@users.noreply.github.com> Date: Thu, 22 Jan 2026 05:19:25 +0100 Subject: [PATCH] [Docs] Remove outdated async_scheduling limitation with speculative decoding (#32775) Signed-off-by: Ifta Khairul Alam Adil Signed-off-by: Ifta khairul Alam Adil <25082512+ikaadil@users.noreply.github.com> --- vllm/config/scheduler.py | 3 --- 1 file changed, 3 deletions(-) diff --git a/vllm/config/scheduler.py b/vllm/config/scheduler.py index 1bceaa933e..5ff9fc930a 100644 --- a/vllm/config/scheduler.py +++ b/vllm/config/scheduler.py @@ -133,9 +133,6 @@ class SchedulerConfig: async_scheduling: bool = Field(default=None) """If set to False, disable async scheduling. Async scheduling helps to avoid gaps in GPU utilization, leading to better latency and throughput. - It is currently not supported with some features such as - speculative decoding and pipeline parallelism, and will be automatically - disabled in those cases. """ stream_interval: int = Field(default=1, ge=1)