@@ -52,6 +52,14 @@ Changelog
     :pr:`123456` by :user:`Joe Bloggs <joeongithub>`.
     where 123456 is the *pull request* number, not the issue number.
 
+:mod:`sklearn.cluster`
+......................
+
+- |API| The `sample_weight` parameter in `predict` for
+  :meth:`cluster.KMeans.predict` and :meth:`cluster.MiniBatchKMeans.predict`
+  is now deprecated and will be removed in v1.5.
+  :pr:`25251` by :user:`Gleb Levitski <glevv>`.
+
 :mod:`sklearn.decomposition`
 ............................
 - |Enhancement| :class:`decomposition.DictionaryLearning` now accepts the parameter
 
@@ -1032,7 +1032,7 @@ def fit_predict(self, X, y=None, sample_weight=None):
         """
         return self.fit(X, sample_weight=sample_weight).labels_
 
-    def predict(self, X, sample_weight=None):
+    def predict(self, X, sample_weight="deprecated"):
         """Predict the closest cluster each sample in X belongs to.
 
         In the vector quantization literature, `cluster_centers_` is called
@@ -1048,6 +1048,10 @@ def predict(self, X, sample_weight=None):
             The weights for each observation in X. If None, all observations
             are assigned equal weight.
 
+            .. deprecated:: 1.3
+               The parameter `sample_weight` is deprecated in version 1.3
+               and will be removed in 1.5.
+
         Returns
         -------
         labels : ndarray of shape (n_samples,)
@@ -1056,7 +1060,15 @@ def predict(self, X, sample_weight=None):
         check_is_fitted(self)
 
         X = self._check_test_data(X)
-        sample_weight = _check_sample_weight(sample_weight, X, dtype=X.dtype)
+        if sample_weight != "deprecated":
+            warnings.warn(
+                "'sample_weight' was deprecated in version 1.3 and "
+                "will be removed in 1.5.",
+                FutureWarning,
+            )
+            sample_weight = _check_sample_weight(sample_weight, X, dtype=X.dtype)
+        else:
+            sample_weight = _check_sample_weight(None, X, dtype=X.dtype)
 
         labels = _labels_inertia_threadpool_limit(
             X,
 
@@ -225,6 +225,19 @@ def test_algorithm_auto_full_deprecation_warning(algorithm):
         assert kmeans._algorithm == "lloyd"
 
 
+@pytest.mark.parametrize("Estimator", [KMeans, MiniBatchKMeans])
+def test_predict_sample_weight_deprecation_warning(Estimator):
+    X = np.random.rand(100, 2)
+    sample_weight = np.random.uniform(size=100)
+    kmeans = Estimator()
+    kmeans.fit(X, sample_weight=sample_weight)
+    warn_msg = (
+        "'sample_weight' was deprecated in version 1.3 and will be removed in 1.5."
+    )
+    with pytest.warns(FutureWarning, match=warn_msg):
+        kmeans.predict(X, sample_weight=sample_weight)
+
+
 def test_minibatch_update_consistency(global_random_seed):
     # Check that dense and sparse minibatch update give the same results
     rng = np.random.RandomState(global_random_seed)