From 5027ed72391e920402136c5ffd5e2962afbccf0a Mon Sep 17 00:00:00 2001
From: gaoziyuan <88373061+gzy19990617@users.noreply.github.com>
Date: Wed, 17 Sep 2025 14:18:31 +0800
Subject: [PATCH] =?UTF-8?q?=E3=80=90BugFif=E3=80=91fix=20ep=20decode=20(#4?=
 =?UTF-8?q?138)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

* suppoort expert num 3 pre rank

* fix ep decode
---
 fastdeploy/model_executor/layers/moe/fused_moe_backend_base.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fastdeploy/model_executor/layers/moe/fused_moe_backend_base.py b/fastdeploy/model_executor/layers/moe/fused_moe_backend_base.py
index 2a54748af..348945dff 100644
--- a/fastdeploy/model_executor/layers/moe/fused_moe_backend_base.py
+++ b/fastdeploy/model_executor/layers/moe/fused_moe_backend_base.py
@@ -166,7 +166,7 @@ class MoEMethodBase(QuantMethodBase):
             else:
                 if layer.fd_config.parallel_config.splitwise_role == "mixed":
                     self.ep_decoder_runner.clean_low_latency_buffer()
-                return self.apply_ep_prefill(layer, x, gate)
+                return self.apply_ep_decode(layer, x, gate)
         else:
             return self.apply_tp(layer, x, gate)