From 981d95b81e210260babe611d771df7d602be024f Mon Sep 17 00:00:00 2001
From: Chi Lo <Chi.Lo@microsoft.com>
Date: Wed, 22 Jan 2025 10:15:16 -0800
Subject: [PATCH] temporarily disable weight/Bias quantization

---
 onnxruntime/core/optimizer/graph_transformer_utils.cc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/onnxruntime/core/optimizer/graph_transformer_utils.cc b/onnxruntime/core/optimizer/graph_transformer_utils.cc
index 8b8d658f06..d9299b89ee 100644
--- a/onnxruntime/core/optimizer/graph_transformer_utils.cc
+++ b/onnxruntime/core/optimizer/graph_transformer_utils.cc
@@ -249,7 +249,7 @@ InlinedVector<std::unique_ptr<GraphTransformer>> GenerateTransformers(
 
       if (!disable_quant_qdq) {
         transformers.emplace_back(std::make_unique<QDQPropagationTransformer>());
-        transformers.emplace_back(std::make_unique<WeightBiasQuantization>());
+        //transformers.emplace_back(std::make_unique<WeightBiasQuantization>());
 
         // EnsureUniqueDQForNodeUnit is actually a required graph transformation. The unique DQ per QDQ node unit input
         // condition that it ensures is important for the partitioning that happens after Level1 optimizers are run.