Statistical Rethinking 2nd Ed, Ch4, Exercise 4H1

It indeed looks like what I am looking for. However, I am trying to replicate the example but if I use the same structure, such as:

with pm.Model(
    coords_mutable={"obs_idx": np.arange(len(d2))}, 
    coords={"parameter": ["intercept", "slope"]}
) as m4_3:

I get TypeError: __init__() got an unexpected keyword argument 'coords_mutable'
I see at the begin of the page that it is used PyMC v5.6.0 which I have updated the package too. Any idea why?