It is unlikely that I will get a specific answer to this question, so I propose to start a discussion on the geometry of the standard model.
The standard approach to this question is that, in addition to space-time, the space of internal symmetries of the Lagrangian is postulated, which determines the equation of motion of an elementary particle. In other words, in the standard approach, the invariance of the group action on the equation of motion is used as an additional internal space.
It would be interesting to know what kind of geometry is hidden behind the standard model of elementary particles. In this regard, let me share my thoughts on the geometric foundation of the Standard Model in this discussion.
The author's interpretation is as follows. Matter moves along the surface of a seven-dimensional sphere. At the same time, the vacuum form of matter is a Clifford torus S3xS3 with three-dimensional spheres moving in the process of evolution in circles so that the radius of one sphere increases and the radius of the corresponding circle decreases, and vice versa, the radius of the second sphere decreases, and the radius of the corresponding circle increases. As a result, the shape of the vacuum is the product of the Clifford torus and the time torus S1xS1. In this case, the Minkowski space-time (with the signature +1,-3) is wound on the product of a three-dimensional sphere of a large radius and a circle of a time torus of a small radius, and the additional (dual) Minkowski space (with the signature +3,-1) is wound on the product of a three-dimensional sphere of a small radius and a circle of a time torus of a large radius. Thus, a doublet of Minkowski spaces, which is an 8-dimensional space with a neutral metric is wound on the product of the Clifford torus and the time torus.
Within the framework of this model, elementary particles are associated with the minimum length closed lines of the product of the Clifford torus and the time torus, and the mass of elementary particles is associated with the pseudo-Euclidean length of the corresponding curve in 8-dimensional space with a neutral metric. Thus, massless particles lie on a compactified isotropic cone. And the most interesting thing in this geometric model is that the group U(1)xSU(2)xSU(3) naturally arises as the symmetry group of the compactified isotropic cone of the doublet of Minkowski spaces.