diff --git a/doc/source/whatsnew/v3.0.0.rst b/doc/source/whatsnew/v3.0.0.rst
index f26c6506477d4..dddf81fa59aff 100644
--- a/doc/source/whatsnew/v3.0.0.rst
+++ b/doc/source/whatsnew/v3.0.0.rst
@@ -122,6 +122,24 @@ These improvements also fixed certain bugs in groupby:
 - :meth:`.DataFrameGroupBy.sum` would have incorrect values when there are multiple groupings, unobserved groups, and non-numeric data (:issue:`43891`)
 - :meth:`.DataFrameGroupBy.value_counts` would produce incorrect results when used with some categorical and some non-categorical groupings and ``observed=False`` (:issue:`56016`)
 
+.. _whatsnew_300.notable_bug_fixes.xs_function_default_level:
+
+`DataFrame.xs()` did not work as expected when level was not specified by default. (:issue:`59098`)
+
+ex.
+.. ipython:: python
+
+    df = pd.DataFrame(dict(i=[1,2,3], j=[1,1,2], x=[10, 100, 1000])).set_index(["i", "j"])
+
+    key = (1, 1)
+
+    # Returns DataFrame as expected:
+    result1 = df.xs(key, drop_level=False, level=list(range(len(key))))
+
+    # Returns Series, but DataFrame was expected:
+    result2 = df.xs(key, drop_level=False)
+
+
 .. _whatsnew_300.notable_bug_fixes.notable_bug_fix2:
 
 notable_bug_fix2
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 0f0078fc3398b..de826c731d159 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
@@ -2,6 +2,7 @@
 from __future__ import annotations
 
 import collections
+import collections.abc
 from copy import deepcopy
 import datetime as dt
 from functools import partial
@@ -4126,7 +4127,12 @@ class   animal   locomotion
             index = self.index
 
         if isinstance(index, MultiIndex):
-            loc, new_index = index._get_loc_level(key, level=0)
+            level = range(len(key)) if isinstance(key, collections.abc.Sequence) else 0
+            loc, new_index = index.get_loc_level(
+                key,
+                level=level,
+                drop_level=drop_level
+            )
             if not drop_level:
                 if lib.is_integer(loc):
                     # Slice index must be an integer or None
diff --git a/pandas/tests/series/indexing/test_xs.py b/pandas/tests/series/indexing/test_xs.py
index a67f3ec708f24..00a008a1a7fbb 100644
--- a/pandas/tests/series/indexing/test_xs.py
+++ b/pandas/tests/series/indexing/test_xs.py
@@ -2,6 +2,7 @@
 import pytest
 
 from pandas import (
+    DataFrame,
     MultiIndex,
     Series,
     date_range,
@@ -80,3 +81,18 @@ def test_xs_key_as_list(self):
 
         with pytest.raises(TypeError, match="list keys are not supported"):
             ser.xs(["a"], axis=0, drop_level=False)
+
+    def test_xs_default_level(self):
+        # GH#59098
+        df = DataFrame(
+            {"i": [1,2,3], "j": [1,1,2], "x": [10, 100, 1000]}
+        ).set_index(["i", "j"])
+        key = (1, 1)
+
+        # Both scenarios should return DataFrame
+        result_with_level = df.xs(key, drop_level=False, level=list(range(len(key))))
+        result_with_default = df.xs(key, drop_level=False)
+
+        assert type(result_with_default) == DataFrame
+        assert type(result_with_level) == type(result_with_default)
+        tm.assert_equal(result_with_level, result_with_default)