diff --git a/bigframes/ml/linear_model.py b/bigframes/ml/linear_model.py
index 32168e9a34..0816ef9b24 100644
--- a/bigframes/ml/linear_model.py
+++ b/bigframes/ml/linear_model.py
@@ -195,7 +195,7 @@ def __init__(
         self,
         *,
         optimize_strategy: Literal[
-            "auto_strategy", "batch_gradient_descent", "normal_equation"
+            "auto_strategy", "batch_gradient_descent"
         ] = "auto_strategy",
         fit_intercept: bool = True,
         l1_reg: Optional[float] = None,
diff --git a/tests/system/large/ml/test_linear_model.py b/tests/system/large/ml/test_linear_model.py
index eaf666fd50..0cc9fc5353 100644
--- a/tests/system/large/ml/test_linear_model.py
+++ b/tests/system/large/ml/test_linear_model.py
@@ -192,8 +192,6 @@ def test_logistic_regression_customized_params_fit_score(
         f"{dataset_id}.temp_configured_logistic_reg_model"
         in reloaded_model._bqml_model.model_name
     )
-    # TODO(garrettwu) optimize_strategy isn't logged in BQML
-    # assert reloaded_model.optimize_strategy == "BATCH_GRADIENT_DESCENT"
     assert reloaded_model.fit_intercept is False
     assert reloaded_model.class_weight == "balanced"
     assert reloaded_model.calculate_p_values is False
diff --git a/third_party/bigframes_vendored/sklearn/linear_model/_logistic.py b/third_party/bigframes_vendored/sklearn/linear_model/_logistic.py
index 49198eb9bd..c52a37018c 100644
--- a/third_party/bigframes_vendored/sklearn/linear_model/_logistic.py
+++ b/third_party/bigframes_vendored/sklearn/linear_model/_logistic.py
@@ -26,8 +26,10 @@ class LogisticRegression(LinearClassifierMixin, BaseEstimator):
     Args:
         optimize_strategy (str, default "auto_strategy"):
             The strategy to train logistic regression models. Possible values are
-            "auto_strategy", "batch_gradient_descent", "normal_equation". Default
-            to "auto_strategy".
+            "auto_strategy" and "batch_gradient_descent". The two are equilevant since
+            "auto_strategy" will fall back to "batch_gradient_descent". The API is kept
+            for consistency.
+            Default to "auto_strategy".
         fit_intercept (default True):
             Default True. Specifies if a constant (a.k.a. bias or intercept)
             should be added to the decision function.