diff --git a/tests/test_cpp_conduit.py b/tests/test_cpp_conduit.py
index 51fcf6936..fce340098 100644
--- a/tests/test_cpp_conduit.py
+++ b/tests/test_cpp_conduit.py
@@ -7,6 +7,7 @@ import exo_planet_pybind11
 import home_planet_very_lonely_traveler
 import pytest
 
+import env  # noqa: F401
 from pybind11_tests import cpp_conduit as home_planet
 
 
@@ -20,6 +21,7 @@ def test_premium_traveler_getattr_actually_exists():
     assert t_h.secret_name == "PremiumTraveler GetAttr: secret_name points: 7"
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should get fixed on GraalPy side")
 def test_call_cpp_conduit_success():
     t_h = home_planet.Traveler("home")
     cap = t_h._pybind11_conduit_v1_(
diff --git a/tests/test_enum.py b/tests/test_enum.py
index 9914b9001..03cd1c1a6 100644
--- a/tests/test_enum.py
+++ b/tests/test_enum.py
@@ -3,9 +3,11 @@ from __future__ import annotations
 
 import pytest
 
+import env  # noqa: F401
 from pybind11_tests import enums as m
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should get fixed on GraalPy side")
 def test_unscoped_enum():
     assert str(m.UnscopedEnum.EOne) == "UnscopedEnum.EOne"
     assert str(m.UnscopedEnum.ETwo) == "UnscopedEnum.ETwo"
@@ -193,6 +195,7 @@ def test_implicit_conversion():
     assert repr(x) == "{<EMode.EFirstMode: 1>: 3, <EMode.ESecondMode: 2>: 4}"
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should get fixed on GraalPy side")
 def test_binary_operators():
     assert int(m.Flags.Read) == 4
     assert int(m.Flags.Write) == 2
diff --git a/tests/test_exceptions.py b/tests/test_exceptions.py
index a8fd105ea..21449d58c 100644
--- a/tests/test_exceptions.py
+++ b/tests/test_exceptions.py
@@ -201,6 +201,7 @@ def test_custom(msg):
     assert msg(excinfo.value) == "[PythonMyException7]: abc"
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should get fixed on GraalPy side")
 def test_nested_throws(capture):
     """Tests nested (e.g. C++ -> Python -> C++) exception handling"""
 
@@ -369,6 +370,7 @@ def _test_flaky_exception_failure_point_init_py_3_12():
     "env.PYPY and sys.version_info[:2] < (3, 12)",
     reason="PyErr_NormalizeException Segmentation fault",
 )
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should be fixed on GraalPy side")
 def test_flaky_exception_failure_point_init():
     if sys.version_info[:2] < (3, 12):
         _test_flaky_exception_failure_point_init_before_py_3_12()
@@ -376,6 +378,7 @@ def test_flaky_exception_failure_point_init():
         _test_flaky_exception_failure_point_init_py_3_12()
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should be fixed on GraalPy side")
 def test_flaky_exception_failure_point_str():
     what, py_err_set_after_what = m.error_already_set_what(
         FlakyException, ("failure_point_str",)
diff --git a/tests/test_gil_scoped.py b/tests/test_gil_scoped.py
index eab92093c..ab5f54f5e 100644
--- a/tests/test_gil_scoped.py
+++ b/tests/test_gil_scoped.py
@@ -211,6 +211,11 @@ def _run_in_threads(test_fn, num_threads, parallel):
 
 @pytest.mark.skipif(sys.platform.startswith("emscripten"), reason="Requires threads")
 @pytest.mark.parametrize("test_fn", ALL_BASIC_TESTS_PLUS_INTENTIONAL_DEADLOCK)
+@pytest.mark.xfail(
+    "env.GRAALPY",
+    reason="GraalPy transiently complains about unfinished threads at process exit",
+    strict=False,
+)
 def test_run_in_process_one_thread(test_fn):
     """Makes sure there is no GIL deadlock when running in a thread.
 
@@ -221,6 +226,11 @@ def test_run_in_process_one_thread(test_fn):
 
 @pytest.mark.skipif(sys.platform.startswith("emscripten"), reason="Requires threads")
 @pytest.mark.parametrize("test_fn", ALL_BASIC_TESTS_PLUS_INTENTIONAL_DEADLOCK)
+@pytest.mark.xfail(
+    "env.GRAALPY",
+    reason="GraalPy transiently complains about unfinished threads at process exit",
+    strict=False,
+)
 def test_run_in_process_multiple_threads_parallel(test_fn):
     """Makes sure there is no GIL deadlock when running in a thread multiple times in parallel.
 
@@ -231,6 +241,11 @@ def test_run_in_process_multiple_threads_parallel(test_fn):
 
 @pytest.mark.skipif(sys.platform.startswith("emscripten"), reason="Requires threads")
 @pytest.mark.parametrize("test_fn", ALL_BASIC_TESTS_PLUS_INTENTIONAL_DEADLOCK)
+@pytest.mark.xfail(
+    "env.GRAALPY",
+    reason="GraalPy transiently complains about unfinished threads at process exit",
+    strict=False,
+)
 def test_run_in_process_multiple_threads_sequential(test_fn):
     """Makes sure there is no GIL deadlock when running in a thread multiple times sequentially.
 
@@ -241,6 +256,11 @@ def test_run_in_process_multiple_threads_sequential(test_fn):
 
 @pytest.mark.skipif(sys.platform.startswith("emscripten"), reason="Requires threads")
 @pytest.mark.parametrize("test_fn", ALL_BASIC_TESTS_PLUS_INTENTIONAL_DEADLOCK)
+@pytest.mark.xfail(
+    "env.GRAALPY",
+    reason="GraalPy transiently complains about unfinished threads at process exit",
+    strict=False,
+)
 def test_run_in_process_direct(test_fn):
     """Makes sure there is no GIL deadlock when using processes.
 
diff --git a/tests/test_operator_overloading.py b/tests/test_operator_overloading.py
index b6760902d..22300eb0f 100644
--- a/tests/test_operator_overloading.py
+++ b/tests/test_operator_overloading.py
@@ -2,10 +2,12 @@ from __future__ import annotations
 
 import pytest
 
+import env  # noqa: F401
 from pybind11_tests import ConstructorStats
 from pybind11_tests import operators as m
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should get fixed on GraalPy side")
 def test_operator_overloading():
     v1 = m.Vector2(1, 2)
     v2 = m.Vector(3, -1)
@@ -83,6 +85,7 @@ def test_operator_overloading():
     assert cstats.move_assignments == 0
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should get fixed on GraalPy side")
 def test_operators_notimplemented():
     """#393: need to return NotSupported to ensure correct arithmetic operator behavior"""
 
diff --git a/tests/test_pytypes.py b/tests/test_pytypes.py
index 77ab583d1..39d0b619b 100644
--- a/tests/test_pytypes.py
+++ b/tests/test_pytypes.py
@@ -262,6 +262,7 @@ def test_str(doc):
         m.str_from_std_string_input,
     ],
 )
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should be fixed on GraalPy side")
 def test_surrogate_pairs_unicode_error(func):
     input_str = "\ud83d\ude4f".encode("utf-8", "surrogatepass")
     with pytest.raises(UnicodeDecodeError):
@@ -420,6 +421,7 @@ def test_accessor_moves():
         pytest.skip("Not defined: PYBIND11_HANDLE_REF_DEBUG")
 
 
+@pytest.mark.xfail("env.GRAALPY", reason="TODO should be fixed on GraalPy side")
 def test_constructors():
     """C++ default and converting constructors are equivalent to type calls in Python"""
     types = [bytes, bytearray, str, bool, int, float, tuple, list, dict, set]