jax.numpy.matmul fails because the shape of vectors doesn't include their components.
import coordinax as cx
q2 = cx.CartesianPos3D.from_([[0.0, 0, 0], [1, 1, 1]], "kpc")
q2.shape
# (2,)
So
import quaxed.numpy as jnp
rot = jnp.asarray([[0., -1, 0], [1, 0, 0], [0, 0, 1])
jnp.matmul(rot, q2)
# TypeError: dot_general requires contracting dimensions to have the same shape, got (3,) and (2,).
jax.numpy.matmulfails because the shape of vectors doesn't include their components.So