Hello again Rick,
I'd like to recap some of what I said in the hidden comment above. To my mind; inflation and geometrization are automatic in an octonion embedding space, because of the built-in dynamism you describe as the octonions 'needing to drive' and I describe as self-evolving or sequentially evolutive properties of non-associative geometries - applied to Physics - detailed somewhat in my essay from last year about the elephant in the room with quantum gravity researchers.
Well it turns out the same elephant is there with cosmologists too. The exact nature of the inflaton remains a mystery, and the question also remains of why it shuts off when it does. What if the inflaton is non-associativity? This makes the octonions the driver of inflation. If we look at the properties of spheres; we find the 7-sphere is maximally spacious in hypersurface area, but not volume. The maximal volume for hyperspheres (or rather balls - filled spheres) is in 5-d.
These parameters determine that in octonionic inflation, things evolve to a 5-d volume; where inflation ends because there is no further to go and things flip into a quaternionic bubble - as in Cartan's rolling ball model of G2, but where the outer ball is assumed to be 4-d. This transition is effected because the change from non-associative to associative geometry makes the previously outward facing force of gravity turn inward - toward the center of massive objects.
I apologize if this is a distraction Rick, but I was trying to solidify these thoughts in my own mind, and it made sense to elaborate on what I said in the comments above.
Warm Regards,
Jonathan