diff --git a/server/text_generation_server/utils/gptq/quant_linear.py b/server/text_generation_server/utils/gptq/quant_linear.py
index f1ba9bf9..b01788bb 100644
--- a/server/text_generation_server/utils/gptq/quant_linear.py
+++ b/server/text_generation_server/utils/gptq/quant_linear.py
@@ -386,7 +386,7 @@ def ext_q4_matmul(x, q4, q4_width):
 
 class Ex4bitLinear:
     """Linear layer implementation with per-group 4-bit quantization of the weights"""
-    def __init__(self, qweight, qzeros, scales, g_idx, bias, bits, groupsize, device, world_size: int):
+    def __init__(self, qweight, qzeros, scales, g_idx, bias, bits, groupsize, device):
         assert bits == 4
 
         self.device = device
@@ -417,9 +417,6 @@ class Ex4bitLinear:
         # Infer groupsize from height of qzeros
         self.groupsize = None
         if self.qzeros.shape[0] > 1:
-            if world_size is None:
-                world_size = 1
-            # self.groupsize = (self.qweight.shape[0] * 8) // (self.qzeros.shape[0] // world_size)
             self.groupsize = (self.qweight.shape[0] * 8) // (self.qzeros.shape[0])
         
         assert groupsize == self.groupsize
diff --git a/server/text_generation_server/utils/layers.py b/server/text_generation_server/utils/layers.py
index dd399ee3..32de46f4 100644
--- a/server/text_generation_server/utils/layers.py
+++ b/server/text_generation_server/utils/layers.py
@@ -156,7 +156,7 @@ def get_linear(weight, bias, quantize, device = None):
                 f"The passed weight is not `gptq` compatible, loader needs to be updated."
             )
 
-        linear = Ex4bitLinear(qweight, qzeros, scales, g_idx, bias, bits, groupsize, device, world_size)
+        linear = Ex4bitLinear(qweight, qzeros, scales, g_idx, bias, bits, groupsize, device)
     else:
         raise NotImplementedError(f"Quantization `{quantize}` is not implemented yet.")
     return linear