diff --git a/server/text_generation_server/cli.py b/server/text_generation_server/cli.py
index 430323bc..5d25bfc5 100644
--- a/server/text_generation_server/cli.py
+++ b/server/text_generation_server/cli.py
@@ -316,7 +316,7 @@ def quantize(
         logger_level=logger_level,
         json_output=json_output,
     )
-    from text_generation_server.utils.gptq.quantize import quantize
+    from text_generation_server.layers.gptq.quantize import quantize
 
     quantize(
         model_id=model_id,
diff --git a/server/text_generation_server/layers/gptq/quantize.py b/server/text_generation_server/layers/gptq/quantize.py
index ca113d8f..8d029817 100644
--- a/server/text_generation_server/layers/gptq/quantize.py
+++ b/server/text_generation_server/layers/gptq/quantize.py
@@ -12,7 +12,7 @@ from huggingface_hub import HfApi
 from accelerate import init_empty_weights
 from text_generation_server.utils import initialize_torch_distributed, Weights
 from text_generation_server.utils.hub import weight_files
-from text_generation_server.utils.gptq.quant_linear import QuantLinear
+from text_generation_server.layers.gptq.quant_linear import QuantLinear
 from loguru import logger
 from typing import Optional