diff --git a/tests/test_dask.py b/tests/test_dask.py
index fa719dbd..ce69d503 100644
--- a/tests/test_dask.py
+++ b/tests/test_dask.py
@@ -1,3 +1,5 @@
+from contextlib import contextmanager
+
 import dask
 import numpy as np
 import pytest
@@ -12,11 +14,11 @@ def xp():
     return array_namespace(da.empty(0))
 
 
-@pytest.fixture
-def no_compute():
+@contextmanager
+def assert_no_compute():
     """
-    Cause the test to raise if at any point anything calls compute() or persist(),
-    e.g. as it can be triggered implicitly by __bool__, __array__, etc.
+    Context manager that raises if at any point inside it anything calls compute()
+    or persist(), e.g. as it can be triggered implicitly by __bool__, __array__, etc.
     """
     def get(dsk, *args, **kwargs):
         raise AssertionError("Called compute() or persist()")
@@ -25,59 +27,68 @@ def get(dsk, *args, **kwargs):
         yield
 
 
-def test_no_compute(no_compute):
-    """Test the no_compute fixture"""
+def test_no_compute():
+    """Test the no_compute context manager"""
     a = da.asarray(True)
     with pytest.raises(AssertionError, match="Called compute"):
-        bool(a)
+        with assert_no_compute():
+            bool(a)
+
+    # Exiting the context manager restores the original scheduler
+    assert bool(a) is True
 
 
 # Test no_compute for functions that use generic _aliases with xp=np
 
-def test_unary_ops_no_compute(xp, no_compute):
-    a = xp.asarray([1.5, -1.5])
-    xp.ceil(a)
-    xp.floor(a)
-    xp.trunc(a)
-    xp.sign(a)
+def test_unary_ops_no_compute(xp):
+    with assert_no_compute():
+        a = xp.asarray([1.5, -1.5])
+        xp.ceil(a)
+        xp.floor(a)
+        xp.trunc(a)
+        xp.sign(a)
 
 
-def test_matmul_tensordot_no_compute(xp, no_compute):
+def test_matmul_tensordot_no_compute(xp):
     A = da.ones((4, 4), chunks=2)
     B = da.zeros((4, 4), chunks=2)
-    xp.matmul(A, B)
-    xp.tensordot(A, B)
+    with assert_no_compute():
+        xp.matmul(A, B)
+        xp.tensordot(A, B)
 
 
 # Test no_compute for functions that are fully bespoke for dask
 
-def test_asarray_no_compute(xp, no_compute):
-    a = xp.arange(10)
-    xp.asarray(a)
-    xp.asarray(a, dtype=np.int16)
-    xp.asarray(a, dtype=a.dtype)
-    xp.asarray(a, copy=True)
-    xp.asarray(a, copy=True, dtype=np.int16)
-    xp.asarray(a, copy=True, dtype=a.dtype)
-    xp.asarray(a, copy=False)
-    xp.asarray(a, copy=False, dtype=a.dtype)
+def test_asarray_no_compute(xp):
+    with assert_no_compute():
+        a = xp.arange(10)
+        xp.asarray(a)
+        xp.asarray(a, dtype=np.int16)
+        xp.asarray(a, dtype=a.dtype)
+        xp.asarray(a, copy=True)
+        xp.asarray(a, copy=True, dtype=np.int16)
+        xp.asarray(a, copy=True, dtype=a.dtype)
+        xp.asarray(a, copy=False)
+        xp.asarray(a, copy=False, dtype=a.dtype)
 
 
 @pytest.mark.parametrize("copy", [True, False])
-def test_astype_no_compute(xp, no_compute, copy):
-    a = xp.arange(10)
-    xp.astype(a, np.int16, copy=copy)
-    xp.astype(a, a.dtype, copy=copy)
+def test_astype_no_compute(xp, copy):
+    with assert_no_compute():
+        a = xp.arange(10)
+        xp.astype(a, np.int16, copy=copy)
+        xp.astype(a, a.dtype, copy=copy)
 
 
-def test_clip_no_compute(xp, no_compute):
-    a = xp.arange(10)
-    xp.clip(a)
-    xp.clip(a, 1)
-    xp.clip(a, 1, 8)
+def test_clip_no_compute(xp):
+    with assert_no_compute():
+        a = xp.arange(10)
+        xp.clip(a)
+        xp.clip(a, 1)
+        xp.clip(a, 1, 8)
 
 
-def test_generators_are_lazy(xp, no_compute):
+def test_generators_are_lazy(xp):
     """
     Test that generator functions are fully lazy, e.g. that
     da.ones(n) is not implemented as da.asarray(np.ones(n))
@@ -85,12 +96,13 @@ def test_generators_are_lazy(xp, no_compute):
     size = 100_000_000_000  # 800 GB
     chunks = size // 10  # 10x 80 GB chunks
 
-    xp.zeros(size, chunks=chunks)
-    xp.ones(size, chunks=chunks)
-    xp.empty(size, chunks=chunks)
-    xp.full(size, fill_value=123, chunks=chunks)
-    a = xp.arange(size, chunks=chunks)
-    xp.zeros_like(a)
-    xp.ones_like(a)
-    xp.empty_like(a)
-    xp.full_like(a, fill_value=123)
+    with assert_no_compute():
+        xp.zeros(size, chunks=chunks)
+        xp.ones(size, chunks=chunks)
+        xp.empty(size, chunks=chunks)
+        xp.full(size, fill_value=123, chunks=chunks)
+        a = xp.arange(size, chunks=chunks)
+        xp.zeros_like(a)
+        xp.ones_like(a)
+        xp.empty_like(a)
+        xp.full_like(a, fill_value=123)