diff --git a/fx2ait/fx2ait/tensor_spec.py b/fx2ait/fx2ait/tensor_spec.py
index 1db415571..6ac2f3a32 100644
--- a/fx2ait/fx2ait/tensor_spec.py
+++ b/fx2ait/fx2ait/tensor_spec.py
@@ -481,6 +481,9 @@ def find_batch_size_dim(cls, inputs: Any) -> []:
             if len(shape) < 2:
                 # By pass for rank-1 tensors. MRS model has rank-1 tensor carry no batch_size info
                 continue
+            if shape[0] == 0:
+                # We expect that 1 is the minimum batch size value anyways.
+                continue
             # Dedup shape value for single tensor
             first_dims.add(shape[0])
             seen_dims = set()
diff --git a/fx2ait/fx2ait/test/test_tensor_spec.py b/fx2ait/fx2ait/test/test_tensor_spec.py
index 4ab33f218..4c7c59118 100644
--- a/fx2ait/fx2ait/test/test_tensor_spec.py
+++ b/fx2ait/fx2ait/test/test_tensor_spec.py
@@ -178,3 +178,21 @@ def test_input_with_no_bs_tensor(self):
             ),
             specs[3],
         )
+
+    def test_input_with_first_dim_zero(self):
+        inputs = [
+            torch.empty([10, 8643], dtype=torch.float16),
+            torch.empty([0, 8643], dtype=torch.float16),
+        ]
+
+        specs = TensorSpec.from_input_list_with_batch_size(inputs, 32)
+
+        self.assertEqual(
+            [
+                TensorSpec(
+                    [IntVar([1, 32], "batch_size"), IntImm(8643)], torch.float16
+                ),
+                TensorSpec([IntImm(0), IntImm(8643)], torch.float16),
+            ],
+            specs,
+        )