diff --git a/.gitignore b/.gitignore
index c6537d0..e82e5bb 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,4 @@
+*.swp
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
diff --git a/napari_animation/_tests/test_interpolation.py b/napari_animation/_tests/test_interpolation.py
index 1ec94e8..c4d62fb 100644
--- a/napari_animation/_tests/test_interpolation.py
+++ b/napari_animation/_tests/test_interpolation.py
@@ -1,3 +1,4 @@
+import numbers
 from dataclasses import asdict
 from typing import NamedTuple
 
@@ -14,17 +15,29 @@
 from napari_animation.interpolation.utils import nested_assert_close
 
 
+def _expected_type(a, b):
+    if isinstance(a, numbers.Integral) and isinstance(b, numbers.Real):
+        return type(b)
+    return type(a)
+
+
 # Actual tests
-@pytest.mark.parametrize("a", [0.0, 0])
-@pytest.mark.parametrize("b", [100.0, 100])
+@pytest.mark.parametrize("a", [0.0, 0, np.float32(0)])
+@pytest.mark.parametrize("b", [100.0, 100, np.float32(100)])
 @pytest.mark.parametrize("fraction", [0, 0.0, 0.5, 1.0, 1])
 def test_interpolate_num(a, b, fraction):
     """Check that interpolation of numbers produces valid output"""
     result = interpolate_num(a, b, fraction)
-    assert isinstance(result, type(a))
+    assert isinstance(result, _expected_type(a, b))
     assert result == fraction * b
 
 
+@pytest.mark.parametrize("b", [1.0, np.float32(1)])
+def test_interpolate_proper_type(b):
+    result = interpolate_num(0, b, 0.5)
+    assert np.isclose(result, 0.5)
+
+
 @pytest.mark.parametrize("a,b", [([0.0, 0.0, 0.0], [1.0, 1.0, 1.0])])
 @pytest.mark.parametrize(
     "fraction,expected",
diff --git a/napari_animation/interpolation/base_interpolation.py b/napari_animation/interpolation/base_interpolation.py
index 2996c00..bb2213f 100644
--- a/napari_animation/interpolation/base_interpolation.py
+++ b/napari_animation/interpolation/base_interpolation.py
@@ -1,4 +1,4 @@
-from numbers import Number
+from numbers import Integral, Number, Real
 from typing import Sequence, Tuple, TypeVar
 
 import numpy as np
@@ -85,6 +85,8 @@ def interpolate_num(a: Number, b: Number, fraction: float) -> Number:
     Interpolated value between a and b at fraction.
     """
     number_cls = type(a)
+    if isinstance(b, Real) and isinstance(a, Integral):
+        number_cls = type(b)
     return number_cls(a + (b - a) * fraction)