diff --git a/docs/src/python/fft.rst b/docs/src/python/fft.rst
index 36d9d78380..78bfe7f8fc 100644
--- a/docs/src/python/fft.rst
+++ b/docs/src/python/fft.rst
@@ -20,5 +20,7 @@ FFT
   irfft2
   rfftn
   irfftn
+  fftfreq
+  rfftfreq
   fftshift
   ifftshift
diff --git a/mlx/fft.cpp b/mlx/fft.cpp
index 6cad15c33b..8ddc1aca46 100644
--- a/mlx/fft.cpp
+++ b/mlx/fft.cpp
@@ -310,4 +310,32 @@ array ifftshift(const array& a, StreamOrDevice s /* = {} */) {
   return ifftshift(a, axes, s);
 }
 
+array fftfreq(int n, double d /* = 1.0 */, StreamOrDevice s /* = {} */) {
+  if (n <= 0) {
+    throw std::invalid_argument("[fftfreq] `n` must be greater than 0.");
+  }
+  if (d == 0.0) {
+    throw std::invalid_argument("[fftfreq] `d` must be non-zero.");
+  }
+  auto pos = arange(0, (n + 1) / 2, float32, s);
+  auto neg = arange(-(n / 2), 0, float32, s);
+  auto freqs = concatenate({pos, neg}, s);
+  auto scale =
+      array(static_cast<float>(1.0 / (static_cast<double>(n) * d)), float32);
+  return multiply(freqs, scale, s);
+}
+
+array rfftfreq(int n, double d /* = 1.0 */, StreamOrDevice s /* = {} */) {
+  if (n <= 0) {
+    throw std::invalid_argument("[rfftfreq] `n` must be greater than 0.");
+  }
+  if (d == 0.0) {
+    throw std::invalid_argument("[rfftfreq] `d` must be non-zero.");
+  }
+  auto freqs = arange(0, n / 2 + 1, float32, s);
+  auto scale =
+      array(static_cast<float>(1.0 / (static_cast<double>(n) * d)), float32);
+  return multiply(freqs, scale, s);
+}
+
 } // namespace mlx::core::fft
diff --git a/mlx/fft.h b/mlx/fft.h
index a36af78de9..f69b8aa2c4 100644
--- a/mlx/fft.h
+++ b/mlx/fft.h
@@ -212,6 +212,13 @@ inline array irfft2(
     StreamOrDevice s = {}) {
   return irfftn(a, axes, norm, s);
 }
+/** Compute the discrete Fourier Transform sample frequencies. */
+MLX_API array fftfreq(int n, double d = 1.0, StreamOrDevice s = {});
+
+/** Compute the discrete Fourier Transform sample frequencies for
+ * `rfft`/`irfft`. */
+MLX_API array rfftfreq(int n, double d = 1.0, StreamOrDevice s = {});
+
 /** Shift the zero-frequency component to the center of the spectrum. */
 MLX_API array fftshift(const array& a, StreamOrDevice s = {});
 
diff --git a/python/src/fft.cpp b/python/src/fft.cpp
index 8333c3f1f4..8c95ffe02d 100644
--- a/python/src/fft.cpp
+++ b/python/src/fft.cpp
@@ -18,6 +18,19 @@ using namespace nb::literals;
 
 namespace {
 
+using IntOrVec = std::variant<int, std::vector<int>>;
+using AxesArg = std::optional<IntOrVec>;
+
+std::optional<std::vector<int>> normalize_axes(const AxesArg& axes) {
+  if (!axes.has_value()) {
+    return std::nullopt;
+  }
+  if (auto axis = std::get_if<int>(&axes.value())) {
+    return std::vector<int>{*axis};
+  }
+  return std::get<std::vector<int>>(axes.value());
+}
+
 mx::fft::FFTNorm parse_norm(std::string_view norm, std::string_view op) {
   if (norm == "backward") {
     return mx::fft::FFTNorm::Backward;
@@ -541,13 +554,52 @@ void init_fft(nb::module_& parent_module) {
         Returns:
             array: The real array containing the inverse of :func:`rfftn`.
       )pbdoc");
+  m.def(
+      "fftfreq",
+      [](int n, double d, mx::StreamOrDevice s) {
+        return mx::fft::fftfreq(n, d, s);
+      },
+      "n"_a,
+      "d"_a = 1.0,
+      "stream"_a = nb::none(),
+      R"pbdoc(
+        Return the discrete Fourier Transform sample frequencies.
+
+        Args:
+            n (int): Window length.
+            d (float, optional): Sample spacing. The default is ``1.0``.
+
+        Returns:
+            array: The sample frequencies as a one-dimensional array of type ``float32``.
+      )pbdoc");
+  m.def(
+      "rfftfreq",
+      [](int n, double d, mx::StreamOrDevice s) {
+        return mx::fft::rfftfreq(n, d, s);
+      },
+      "n"_a,
+      "d"_a = 1.0,
+      "stream"_a = nb::none(),
+      R"pbdoc(
+        Return the discrete Fourier Transform sample frequencies
+        for use with :func:`rfft` and :func:`irfft`.
+
+        The returned array contains the non-negative frequency terms
+        in the range ``[0, floor(n/2)]``.
+
+        Args:
+            n (int): Window length.
+            d (float, optional): Sample spacing. The default is ``1.0``.
+
+        Returns:
+            array: The sample frequencies as a one-dimensional array of type ``float32``.
+      )pbdoc");
   m.def(
       "fftshift",
-      [](const mx::array& a,
-         const std::optional<std::vector<int>>& axes,
-         mx::StreamOrDevice s) {
-        if (axes.has_value()) {
-          return mx::fft::fftshift(a, axes.value(), s);
+      [](const mx::array& a, const AxesArg& axes, mx::StreamOrDevice s) {
+        auto normalized_axes = normalize_axes(axes);
+        if (normalized_axes.has_value()) {
+          return mx::fft::fftshift(a, normalized_axes.value(), s);
         } else {
           return mx::fft::fftshift(a, s);
         }
@@ -560,7 +612,7 @@ void init_fft(nb::module_& parent_module) {
 
         Args:
             a (array): The input array.
-            axes (list(int), optional): Axes over which to perform the shift.
+            axes (int or list(int), optional): Axis or axes over which to perform the shift.
                If ``None``, shift all axes. 
 
         Returns:
@@ -568,11 +620,10 @@ void init_fft(nb::module_& parent_module) {
       )pbdoc");
   m.def(
       "ifftshift",
-      [](const mx::array& a,
-         const std::optional<std::vector<int>>& axes,
-         mx::StreamOrDevice s) {
-        if (axes.has_value()) {
-          return mx::fft::ifftshift(a, axes.value(), s);
+      [](const mx::array& a, const AxesArg& axes, mx::StreamOrDevice s) {
+        auto normalized_axes = normalize_axes(axes);
+        if (normalized_axes.has_value()) {
+          return mx::fft::ifftshift(a, normalized_axes.value(), s);
         } else {
           return mx::fft::ifftshift(a, s);
         }
@@ -586,7 +637,7 @@ void init_fft(nb::module_& parent_module) {
 
         Args:
             a (array): The input array.
-            axes (list(int), optional): Axes over which to perform the inverse shift.
+            axes (int or list(int), optional): Axis or axes over which to perform the inverse shift.
                If ``None``, shift all axes. 
 
         Returns:
diff --git a/python/tests/test_fft.py b/python/tests/test_fft.py
index 23aaf27af0..9c03530b04 100644
--- a/python/tests/test_fft.py
+++ b/python/tests/test_fft.py
@@ -246,6 +246,48 @@ def test_fft_throws(self):
         with self.assertRaises(ValueError):
             mx.fft.fft(mx.array([1.0]), norm="invalid")
 
+    def test_fftfreq(self):
+        for n, d in [(1, 1.0), (4, 0.5), (5, 0.25), (8, -0.5), (6, 1.0)]:
+            out = mx.fft.fftfreq(n, d=d)
+            expected = np.fft.fftfreq(n, d=d).astype(np.float32)
+            self.assertEqual(out.dtype, mx.float32)
+            np.testing.assert_allclose(out, expected, atol=0.0, rtol=0.0)
+
+        with self.assertRaises(ValueError):
+            mx.fft.fftfreq(0)
+
+        with self.assertRaises(ValueError):
+            mx.fft.fftfreq(-1)
+
+        # Test default d=1.0
+        out = mx.fft.fftfreq(8)
+        expected = np.fft.fftfreq(8).astype(np.float32)
+        np.testing.assert_allclose(out, expected, atol=0.0, rtol=0.0)
+
+        with self.assertRaises(ValueError):
+            mx.fft.fftfreq(4, d=0.0)
+
+    def test_rfftfreq(self):
+        for n, d in [(1, 1.0), (4, 0.5), (5, 0.25), (8, -0.5), (6, 1.0)]:
+            out = mx.fft.rfftfreq(n, d=d)
+            expected = np.fft.rfftfreq(n, d=d).astype(np.float32)
+            self.assertEqual(out.dtype, mx.float32)
+            np.testing.assert_allclose(out, expected, atol=0.0, rtol=0.0)
+
+        # Test default d=1.0
+        out = mx.fft.rfftfreq(8)
+        expected = np.fft.rfftfreq(8).astype(np.float32)
+        np.testing.assert_allclose(out, expected, atol=0.0, rtol=0.0)
+
+        with self.assertRaises(ValueError):
+            mx.fft.rfftfreq(0)
+
+        with self.assertRaises(ValueError):
+            mx.fft.rfftfreq(-1)
+
+        with self.assertRaises(ValueError):
+            mx.fft.rfftfreq(4, d=0.0)
+
     def test_fftshift(self):
         # Test 1D arrays
         r = np.random.rand(100).astype(np.float32)
@@ -253,11 +295,14 @@ def test_fftshift(self):
 
         # Test with specific axis
         r = np.random.rand(4, 6).astype(np.float32)
+        self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=0)
+        self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=1)
         self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=[0])
         self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=[1])
         self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=[0, 1])
 
         # Test with negative axes
+        self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=-1)
         self.check_mx_np(mx.fft.fftshift, np.fft.fftshift, r, axes=[-1])
 
         # Test with odd lengths
@@ -278,11 +323,14 @@ def test_ifftshift(self):
 
         # Test with specific axis
         r = np.random.rand(4, 6).astype(np.float32)
+        self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=0)
+        self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=1)
         self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=[0])
         self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=[1])
         self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=[0, 1])
 
         # Test with negative axes
+        self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=-1)
         self.check_mx_np(mx.fft.ifftshift, np.fft.ifftshift, r, axes=[-1])
 
         # Test with odd lengths