From d79d9f078079e4108ec9d28b261a8e002e763338 Mon Sep 17 00:00:00 2001 From: Wentao Ye <44945378+yewentao256@users.noreply.github.com> Date: Wed, 5 Nov 2025 20:03:09 -0500 Subject: [PATCH] [Bug] Fix cpu disable shared_experts `VLLM_DISABLE_SHARED_EXPERTS_STREAM` (#28157) Signed-off-by: yewentao256 --- vllm/platforms/cpu.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/platforms/cpu.py b/vllm/platforms/cpu.py index 4b9f4aef0..60a8e90cd 100644 --- a/vllm/platforms/cpu.py +++ b/vllm/platforms/cpu.py @@ -299,7 +299,7 @@ class CpuPlatform(Platform): os.environ["TORCHINDUCTOR_COMPILE_THREADS"] = "1" # Disable multi-stream for shared experts as no Stream on CPU - os.environ["VLLM_DISABLE_SHARED_EXPERTS_STREAM"] = "0" + os.environ["VLLM_DISABLE_SHARED_EXPERTS_STREAM"] = "1" # Intel OpenMP setting ld_preload_str = os.getenv("LD_PRELOAD", "")