diff --git a/doc/whats_new/upcoming_changes/sklearn.linear_model/31888.api.rst b/doc/whats_new/upcoming_changes/sklearn.linear_model/31888.api.rst
new file mode 100644
index 0000000000000..a1ac21999bb09
--- /dev/null
+++ b/doc/whats_new/upcoming_changes/sklearn.linear_model/31888.api.rst
@@ -0,0 +1,4 @@
+- Raising error in :class:`sklearn.linear_model.LogisticRegression` when
+  liblinear solver is used and input X values are larger than 1e30,
+  the liblinear solver freezes otherwise.
+  By :user:`Shruti Nath <snath-xoc>`.
diff --git a/sklearn/linear_model/_logistic.py b/sklearn/linear_model/_logistic.py
index f921f473da835..c803bdc0ba72d 100644
--- a/sklearn/linear_model/_logistic.py
+++ b/sklearn/linear_model/_logistic.py
@@ -1295,6 +1295,12 @@ def fit(self, X, y, sample_weight=None):
         multi_class = _check_multi_class(multi_class, solver, len(self.classes_))
 
         if solver == "liblinear":
+            if np.max(X) > 1e30:
+                raise ValueError(
+                    "Using the 'liblinear' solver while X contains a maximum "
+                    "value > 1e30 results in a frozen fit. Please choose another "
+                    "solver or rescale the input X."
+                )
             if len(self.classes_) > 2:
                 warnings.warn(
                     "Using the 'liblinear' solver for multiclass classification is "
diff --git a/sklearn/linear_model/tests/test_logistic.py b/sklearn/linear_model/tests/test_logistic.py
index 6b08be5a95a0d..440c5d4fdbbb0 100644
--- a/sklearn/linear_model/tests/test_logistic.py
+++ b/sklearn/linear_model/tests/test_logistic.py
@@ -2319,6 +2319,23 @@ def test_large_sparse_matrix(solver, global_random_seed, csr_container):
         LogisticRegression(solver=solver).fit(X, y)
 
 
+def test_liblinear_with_large_values():
+    # Liblinear freezes when X.max() ~ 1e100, see issue #7486.
+    # We preemptively raise an error when X.max() > 1e30.
+
+    # generate sparse matrix with int64 indices
+    X = np.array([0, 1e100]).reshape(-1, 1)
+    y = np.array([0, 1])
+
+    msg = (
+        "Using the 'liblinear' solver while X contains a maximum "
+        "value > 1e30 results in a frozen fit. Please choose another "
+        "solver or rescale the input X."
+    )
+    with pytest.raises(ValueError, match=msg):
+        LogisticRegression(solver="liblinear").fit(X, y)
+
+
 def test_single_feature_newton_cg():
     # Test that Newton-CG works with a single feature and intercept.
     # Non-regression test for issue #23605.