From 568ba7216a4a0de632e6a2660601ae6d08ee9cb1 Mon Sep 17 00:00:00 2001
From: Cheng Wan <54331508+ch-wan@users.noreply.github.com>
Date: Sun, 17 May 2026 02:20:14 -0700
Subject: [PATCH] Fix logging for inplace setting in the flashInfer-trtllm
 backend (#25522)

---
 python/sglang/srt/layers/moe/fused_moe_triton/layer.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/python/sglang/srt/layers/moe/fused_moe_triton/layer.py b/python/sglang/srt/layers/moe/fused_moe_triton/layer.py
index 9b8431950..d8fc286fb 100644
--- a/python/sglang/srt/layers/moe/fused_moe_triton/layer.py
+++ b/python/sglang/srt/layers/moe/fused_moe_triton/layer.py
@@ -70,6 +70,7 @@ from sglang.srt.utils import (
     get_bool_env_var,
     is_cpu,
     is_hip,
+    print_info_once,
     round_up,
 )
 from sglang.srt.utils.custom_op import register_custom_op
@@ -313,7 +314,7 @@ class FusedMoE(torch.nn.Module):
             or get_moe_runner_backend().is_flashinfer_trtllm()
         ):
             if self.moe_runner_config.inplace:
-                logging.info(
+                print_info_once(
                     "Setting inplace to False for FlashInfer TRTLLM MoE backend."
                 )
             self.moe_runner_config.inplace = False