From 1f43c121d57a1fdb8915761450769b8b74dc4e56 Mon Sep 17 00:00:00 2001 From: Vaibhav Sourirajan <113937608+vsourirajan@users.noreply.github.com> Date: Fri, 2 Jan 2026 00:13:36 -0500 Subject: [PATCH] Remove unused `use_marlin` variable in `Mxfp4MoEMethod` (#31549) Signed-off-by: vaibhav sourirajan --- vllm/model_executor/layers/quantization/mxfp4.py | 1 - 1 file changed, 1 deletion(-) diff --git a/vllm/model_executor/layers/quantization/mxfp4.py b/vllm/model_executor/layers/quantization/mxfp4.py index dc0fbfa7d..4fabb426b 100644 --- a/vllm/model_executor/layers/quantization/mxfp4.py +++ b/vllm/model_executor/layers/quantization/mxfp4.py @@ -240,7 +240,6 @@ class Mxfp4MoEMethod(FusedMoEMethodBase): self.mxfp4_backend = get_mxfp4_backend(moe.is_lora_enabled) self.marlin_input_dtype = None - self.use_marlin = self.mxfp4_backend == Mxfp4Backend.MARLIN self.max_capture_size = ( get_current_vllm_config().compilation_config.max_cudagraph_capture_size )