From 1e55dfa7e552e0995630a2563aeae443945e2e81 Mon Sep 17 00:00:00 2001
From: JartX <sagformas@gmail.com>
Date: Mon, 11 Aug 2025 09:13:30 +0200
Subject: [PATCH] [BUGFIX] KeyError 'layers.14.mlp.gate.g_idx' for Qwen3-MoE
 with GPTQ on ROCm (#22017)

---
 vllm/model_executor/models/qwen3_moe.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/vllm/model_executor/models/qwen3_moe.py b/vllm/model_executor/models/qwen3_moe.py
index 3d1e72299b4b8..9b49952f37244 100644
--- a/vllm/model_executor/models/qwen3_moe.py
+++ b/vllm/model_executor/models/qwen3_moe.py
@@ -149,7 +149,7 @@ class Qwen3MoeSparseMoeBlock(nn.Module):
         self.gate = ReplicatedLinear(config.hidden_size,
                                      config.num_experts,
                                      bias=False,
-                                     quant_config=None,
+                                     quant_config=quant_config,
                                      prefix=f"{prefix}.gate")
 
     def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: