From 059779231f158b8b570e71aaa5c66f49b41b2fb1 Mon Sep 17 00:00:00 2001 From: Frank Wang <41319051+frankwang28@users.noreply.github.com> Date: Thu, 19 Feb 2026 22:07:57 -0800 Subject: [PATCH] [Minor] Add logging when using MXFP4 MXFP8 TRTLLM backend (#34916) Signed-off-by: frankwang28 Signed-off-by: Frank Wang <41319051+frankwang28@users.noreply.github.com> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com> --- vllm/model_executor/layers/quantization/mxfp4.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/vllm/model_executor/layers/quantization/mxfp4.py b/vllm/model_executor/layers/quantization/mxfp4.py index 4b24885b43..492963855e 100644 --- a/vllm/model_executor/layers/quantization/mxfp4.py +++ b/vllm/model_executor/layers/quantization/mxfp4.py @@ -128,6 +128,9 @@ def get_mxfp4_backend(with_lora_support: bool) -> Mxfp4Backend: and has_flashinfer() and envs.VLLM_USE_FLASHINFER_MOE_MXFP4_MXFP8 ): + logger.info_once( + "Using FlashInfer MXFP4 MXFP8 TRTLLM backend for SM100", scope="local" + ) return Mxfp4Backend.SM100_FI_MXFP4_MXFP8_TRTLLM elif current_platform.is_device_capability_family(100) and has_flashinfer(): logger.info_once(