diff --git a/src/galax/potential/_potential/builtin.py b/src/galax/potential/_potential/builtin.py
index b6be3dea..72f5a13e 100644
--- a/src/galax/potential/_potential/builtin.py
+++ b/src/galax/potential/_potential/builtin.py
@@ -1,12 +1,12 @@
 """galax: Galactic Dynamix in Jax."""
 
 __all__ = [
-    "KeplerPotential",
-    "MiyamotoNagaiPotential",
-    "NullPotential",
     "BarPotential",
     "IsochronePotential",
+    "KeplerPotential",
+    "MiyamotoNagaiPotential",
     "NFWPotential",
+    "NullPotential",
 ]
 
 from dataclasses import KW_ONLY
@@ -35,60 +35,6 @@
 # -------------------------------------------------------------------
 
 
-class KeplerPotential(AbstractPotential):
-    r"""The Kepler potential for a point mass.
-
-    .. math::
-        \Phi = -\frac{G M(t)}{r}
-    """
-
-    m: AbstractParameter = ParameterField(dimensions=mass)  # type: ignore[assignment]
-
-    @partial_jit()
-    def _potential_energy(
-        self, q: BatchVec3, /, t: BatchableFloatOrIntScalarLike
-    ) -> BatchFloatScalar:
-        r = xp.linalg.norm(q, axis=-1)
-        return -self._G * self.m(t) / r
-
-
-# -------------------------------------------------------------------
-
-
-class MiyamotoNagaiPotential(AbstractPotential):
-    m: AbstractParameter = ParameterField(dimensions=mass)  # type: ignore[assignment]
-    a: AbstractParameter = ParameterField(dimensions=length)  # type: ignore[assignment]
-    b: AbstractParameter = ParameterField(dimensions=length)  # type: ignore[assignment]
-
-    @partial_jit()
-    def _potential_energy(
-        self, q: BatchVec3, /, t: BatchableFloatOrIntScalarLike
-    ) -> BatchFloatScalar:
-        x, y, z = q[..., 0], q[..., 1], q[..., 2]
-        R2 = x**2 + y**2
-        return (
-            -self._G
-            * self.m(t)
-            / xp.sqrt(R2 + xp.square(xp.sqrt(z**2 + self.b(t) ** 2) + self.a(t)))
-        )
-
-
-# -------------------------------------------------------------------
-
-
-class NullPotential(AbstractPotential):
-    """Null potential, i.e. no potential."""
-
-    @partial_jit()
-    def _potential_energy(
-        self, q: BatchVec3, /, t: BatchableFloatOrIntScalarLike
-    ) -> BatchFloatScalar:
-        return xp.zeros(q.shape[:-1], dtype=q.dtype)
-
-
-# -------------------------------------------------------------------
-
-
 class BarPotential(AbstractPotential):
     """Rotating bar potentil, with hard-coded rotation.
 
@@ -155,6 +101,47 @@ def _potential_energy(
 # -------------------------------------------------------------------
 
 
+class KeplerPotential(AbstractPotential):
+    r"""The Kepler potential for a point mass.
+
+    .. math::
+        \Phi = -\frac{G M(t)}{r}
+    """
+
+    m: AbstractParameter = ParameterField(dimensions=mass)  # type: ignore[assignment]
+
+    @partial_jit()
+    def _potential_energy(
+        self, q: BatchVec3, /, t: BatchableFloatOrIntScalarLike
+    ) -> BatchFloatScalar:
+        r = xp.linalg.norm(q, axis=-1)
+        return -self._G * self.m(t) / r
+
+
+# -------------------------------------------------------------------
+
+
+class MiyamotoNagaiPotential(AbstractPotential):
+    m: AbstractParameter = ParameterField(dimensions=mass)  # type: ignore[assignment]
+    a: AbstractParameter = ParameterField(dimensions=length)  # type: ignore[assignment]
+    b: AbstractParameter = ParameterField(dimensions=length)  # type: ignore[assignment]
+
+    @partial_jit()
+    def _potential_energy(
+        self, q: BatchVec3, /, t: BatchableFloatOrIntScalarLike
+    ) -> BatchFloatScalar:
+        x, y, z = q[..., 0], q[..., 1], q[..., 2]
+        R2 = x**2 + y**2
+        return (
+            -self._G
+            * self.m(t)
+            / xp.sqrt(R2 + xp.square(xp.sqrt(z**2 + self.b(t) ** 2) + self.a(t)))
+        )
+
+
+# -------------------------------------------------------------------
+
+
 class NFWPotential(AbstractPotential):
     """NFW Potential."""
 
@@ -171,3 +158,16 @@ def _potential_energy(
         r2 = q[..., 0] ** 2 + q[..., 1] ** 2 + q[..., 2] ** 2
         m = xp.sqrt(r2 + self.softening_length) / self.r_s(t)
         return v_h2 * xp.log(1.0 + m) / m
+
+
+# -------------------------------------------------------------------
+
+
+class NullPotential(AbstractPotential):
+    """Null potential, i.e. no potential."""
+
+    @partial_jit()
+    def _potential_energy(
+        self, q: BatchVec3, /, t: BatchableFloatOrIntScalarLike
+    ) -> BatchFloatScalar:
+        return xp.zeros(q.shape[:-1], dtype=q.dtype)