diff --git a/doc/whats_new/upcoming_changes/sklearn.neural_network/24788.fix.rst b/doc/whats_new/upcoming_changes/sklearn.neural_network/24788.fix.rst
new file mode 100644
index 0000000000000..ea67942daec59
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.neural_network/24788.fix.rst
@@ -0,0 +1,3 @@
+:class:`neural_network.MLPRegressor` now raises an informative error when
+`early_stopping` is set and the computed validation set is too small.
+By :user:`David Shumway <davidshumway>`.
diff --git a/sklearn/neural_network/_multilayer_perceptron.py b/sklearn/neural_network/_multilayer_perceptron.py
index 51ff4176a0524..d18f873e8a0db 100644
--- a/sklearn/neural_network/_multilayer_perceptron.py
+++ b/sklearn/neural_network/_multilayer_perceptron.py
@@ -668,6 +668,12 @@ def _fit_stochastic(
                     test_size=self.validation_fraction,
                     stratify=stratify,
                 )
+            if X_val.shape[0] < 2:
+                raise ValueError(
+                    "The validation set is too small. Increase 'validation_fraction' "
+                    "or the size of your dataset."
+                )
+
             if is_classifier(self):
                 y_val = self._label_binarizer.inverse_transform(y_val)
         else:
diff --git a/sklearn/neural_network/tests/test_mlp.py b/sklearn/neural_network/tests/test_mlp.py
index f788426ad60d2..417d15b0f6cf2 100644
--- a/sklearn/neural_network/tests/test_mlp.py
+++ b/sklearn/neural_network/tests/test_mlp.py
@@ -1084,3 +1084,11 @@ def test_mlp_vs_poisson_glm_equivalent(global_random_seed):
         random_state=np.random.RandomState(global_random_seed + 1),
     ).fit(X, y)
     assert not np.allclose(mlp.predict(X), glm.predict(X), rtol=1e-4)
+
+
+def test_minimum_input_sample_size():
+    """Check error message when the validation set is too small."""
+    X, y = make_regression(n_samples=2, n_features=5, random_state=0)
+    model = MLPRegressor(early_stopping=True, random_state=0)
+    with pytest.raises(ValueError, match="The validation set is too small"):
+        model.fit(X, y)