My occasional posts on ‘Ockham's razor’ are knocked into a cocked hat by the more systematic research of the idiosyncratic Scotus scholar Antonie Vos, whose massive book The Philosophy of John Duns Scotus (2006) includes a section entitled ‘Methodological parsimony: the razor Scoti’ (§8.2). This should be the first port of call for anyone researching the history of the principle.
Here are the formulations quoted by Vos on pp. 304f.:
Numquam est ponenda pluralitas sine necessitate (QM I.4)
Pluralitas numquam ponenda est sine necessitate (IV.2)
Numquam ponenda sunt plura sine necessitate (VII.12)
Pluralitas non est ponenda sine necessitate (VII.12)
Nihil non manifestum ponendum est a philosophantibus sine necessitate (VII.18)
Non est ponenda pluralitas entium sine ratione (Lect. II.2)
Plura non sunt ponenda sine necessitate (II.14)
Sed haec positio ponere videtur pluralitatem sine necessitate (Ord. II.2)
Pluralitas specierum non videtur ponenda sine necessitate manifesta (III.34)
Vos points further to Lectura I.2.202 and Ordinatio IV.11.3 and IV.11.14, which I'm afraid I don't have to hand.