From c42f5285ec9851dacfa370e163e7837e87fd7b6d Mon Sep 17 00:00:00 2001 From: Tony Lin Date: Tue, 22 Sep 2026 23:38:04 +0800 Subject: [PATCH] [Bugfix] prioritize architecture capability before DeepGEMM availability check (#58073) Signed-off-by: Tony Lin --- vllm/utils/deep_gemm.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/utils/deep_gemm.py b/vllm/utils/deep_gemm.py index 2a0d4eee4a..0eb979ce05 100644 --- a/vllm/utils/deep_gemm.py +++ b/vllm/utils/deep_gemm.py @@ -112,7 +112,7 @@ def is_deep_gemm_supported() -> bool: Currently, only Hopper and Blackwell GPUs are supported. """ is_supported_arch = current_platform.support_deep_gemm() - return envs.VLLM_USE_DEEP_GEMM and has_deep_gemm() and is_supported_arch + return is_supported_arch and envs.VLLM_USE_DEEP_GEMM and has_deep_gemm() @functools.cache