diff --git a/GEMINI.md b/GEMINI.md
deleted file mode 100644
index 4de5912527..0000000000
--- a/GEMINI.md
+++ /dev/null
@@ -1,5 +0,0 @@
-# Contribution guidelines, tailored for LLM agents
-
-@.gemini/common/docs.md
-
-@.gemini/common/constraints.md
diff --git a/bigframes/bigquery/_operations/ml.py b/bigframes/bigquery/_operations/ml.py
index d5b1786b25..3e5d6fb263 100644
--- a/bigframes/bigquery/_operations/ml.py
+++ b/bigframes/bigquery/_operations/ml.py
@@ -480,6 +480,39 @@ def generate_text(
         return session.read_gbq_query(sql)
 
 
+@log_adapter.method_logger(custom_base_name="bigquery_ml")
+def get_insights(
+    model: Union[bigframes.ml.base.BaseEstimator, str, pd.Series],
+) -> dataframe.DataFrame:
+    """
+    Gets insights from a BigQuery ML model.
+
+    See the `BigQuery ML GET_INSIGHTS function syntax
+    <https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-get-insights>`_
+    for additional reference.
+
+    Args:
+        model (bigframes.ml.base.BaseEstimator, str, or pd.Series):
+            The model to get insights from.
+
+    Returns:
+        bigframes.pandas.DataFrame:
+            The insights.
+    """
+    import bigframes.pandas as bpd
+
+    model_name, session = utils.get_model_name_and_session(model)
+
+    sql = bigframes.core.sql.ml.get_insights(
+        model_name=model_name,
+    )
+
+    if session is None:
+        return bpd.read_gbq_query(sql)
+    else:
+        return session.read_gbq_query(sql)
+
+
 @log_adapter.method_logger(custom_base_name="bigquery_ml")
 def generate_embedding(
     model: Union[bigframes.ml.base.BaseEstimator, str, pd.Series],
diff --git a/bigframes/bigquery/ml.py b/bigframes/bigquery/ml.py
index b1b33d0dbd..9b0d77d5b8 100644
--- a/bigframes/bigquery/ml.py
+++ b/bigframes/bigquery/ml.py
@@ -25,6 +25,7 @@
     explain_predict,
     generate_embedding,
     generate_text,
+    get_insights,
     global_explain,
     predict,
     transform,
@@ -39,4 +40,5 @@
     "transform",
     "generate_text",
     "generate_embedding",
+    "get_insights",
 ]
diff --git a/bigframes/core/sql/ml.py b/bigframes/core/sql/ml.py
index 391d905d2f..0edb784c37 100644
--- a/bigframes/core/sql/ml.py
+++ b/bigframes/core/sql/ml.py
@@ -266,6 +266,16 @@ def generate_text(
     return sql
 
 
+def get_insights(
+    model_name: str,
+) -> str:
+    """Encode the ML.GET_INSIGHTS statement.
+    See https://cloud.google.com/bigquery/docs/reference/standard-sql/bigqueryml-syntax-get-insights for reference.
+    """
+    sql = f"SELECT * FROM ML.GET_INSIGHTS(MODEL {sg_sql.to_sql(sg_sql.identifier(model_name))})\n"
+    return sql
+
+
 def generate_embedding(
     model_name: str,
     table: str,
diff --git a/tests/system/large/bigquery/test_ml.py b/tests/system/large/bigquery/test_ml.py
index 20a62ae2b6..f0f7d4f691 100644
--- a/tests/system/large/bigquery/test_ml.py
+++ b/tests/system/large/bigquery/test_ml.py
@@ -64,6 +64,32 @@ def test_generate_embedding_with_options(embedding_model):
     assert len(embedding[0]) == 256
 
 
+def test_get_insights(dataset_id):
+    df = bpd.DataFrame(
+        {
+            "dim1": ["a", "a", "b", "b", "a", "a", "b", "b"],
+            "dim2": ["x", "y", "x", "y", "x", "y", "x", "y"],
+            "metric": [10, 20, 30, 40, 12, 25, 35, 45],
+            "is_test": [False, False, False, False, True, True, True, True],
+        }
+    )
+    model_name = f"{dataset_id}.contribution_analysis_model"
+
+    ml.create_model(
+        model_name=model_name,
+        options={
+            "model_type": "CONTRIBUTION_ANALYSIS",
+            "contribution_metric": "SUM(metric)",
+            "is_test_col": "is_test",
+        },
+        training_data=df,
+    )
+
+    result = ml.get_insights(model_name)
+    assert len(result) > 0
+    assert "contributors" in result.columns
+
+
 def test_create_model_linear_regression(dataset_id):
     df = bpd.DataFrame({"x": [1, 2, 3], "y": [2, 4, 6]})
     model_name = f"{dataset_id}.linear_regression_model"
diff --git a/tests/unit/bigquery/test_ml.py b/tests/unit/bigquery/test_ml.py
index e5c957767b..6d39901a35 100644
--- a/tests/unit/bigquery/test_ml.py
+++ b/tests/unit/bigquery/test_ml.py
@@ -177,6 +177,15 @@ def test_generate_text_with_pandas_dataframe(read_pandas_mock, read_gbq_query_mo
     assert "'TYPE' AS request_type" in generated_sql
 
 
+@mock.patch("bigframes.pandas.read_gbq_query")
+def test_get_insights(read_gbq_query_mock):
+    ml_ops.get_insights(MODEL_SERIES)
+    read_gbq_query_mock.assert_called_once()
+    generated_sql = read_gbq_query_mock.call_args[0][0]
+    assert "ML.GET_INSIGHTS" in generated_sql
+    assert f"MODEL `{MODEL_NAME}`" in generated_sql
+
+
 @mock.patch("bigframes.pandas.read_gbq_query")
 @mock.patch("bigframes.pandas.read_pandas")
 def test_generate_embedding_with_pandas_dataframe(
diff --git a/tests/unit/core/sql/snapshots/test_ml/test_get_insights_model_basic/get_insights_model_basic.sql b/tests/unit/core/sql/snapshots/test_ml/test_get_insights_model_basic/get_insights_model_basic.sql
new file mode 100644
index 0000000000..a3f2680c17
--- /dev/null
+++ b/tests/unit/core/sql/snapshots/test_ml/test_get_insights_model_basic/get_insights_model_basic.sql
@@ -0,0 +1 @@
+SELECT * FROM ML.GET_INSIGHTS(MODEL `my_project.my_dataset.my_model`)
diff --git a/tests/unit/core/sql/test_ml.py b/tests/unit/core/sql/test_ml.py
index 27b7a00ac2..bb3b61a949 100644
--- a/tests/unit/core/sql/test_ml.py
+++ b/tests/unit/core/sql/test_ml.py
@@ -203,6 +203,13 @@ def test_generate_text_model_with_options(snapshot):
     snapshot.assert_match(sql, "generate_text_model_with_options.sql")
 
 
+def test_get_insights_model_basic(snapshot):
+    sql = bigframes.core.sql.ml.get_insights(
+        model_name="my_project.my_dataset.my_model",
+    )
+    snapshot.assert_match(sql, "get_insights_model_basic.sql")
+
+
 def test_generate_embedding_model_basic(snapshot):
     sql = bigframes.core.sql.ml.generate_embedding(
         model_name="my_project.my_dataset.my_model",