Usually, stepper motor windings are customized for an application. In other words, it is NOT normal to use off-the-shelf stepper motor designs in a mass-produced product.
Given a ~fixed volume of copper, the tradeoff space is wire gauge vs number of windings → winding current, resistance, inductance, which affects how quickly the motor coils will charge and discharge, and average torque output at high speeds.
Typical lead-times for prototype custom motors are 2-4wks.
As you scale to production, you go from a prototype manufacturing line to [domestic manufacturing, depending on vendor] to overseas mfg at high volumes.
In general, stepper motors are commodity items governed by fundamentals:

HOWEVER, there is still innovation in stepper motor design, along dimensions such as:

These enable some motors that yield better performance in the same package size, eg Lin Engineering G4518 Series, 98ozf-in holding torque for G4518L, vs 83ozf-in for 4118L.

Motors are rated in amps/phase, which is an RMS value, such that the total power dissipated by the motor (sum of both phases) is constant at any point in the microstepping commutation table.
Driver current is set as PEAK amps, which is ~1.4x amps RMS.
A common mistake is to equate these two, and set your drive current too low.
Caveat: if you park your motor at full-current in a position where only one phase is on, it can in theory overheat the winding, but in practice I haven’t seen indications of this being a real problem. Some drivers support lower current levels when stopped.

The motors can represent a significant portion of the overall system compliance, especially in low-ratio (e.g. belt-driven) mechanisms.
From Acarnley (p. 26):
Peak stiffness is when the electrical and mechanical positions of the motor are in phase.
Maximum torque output occurs at +/- 1 step. Deflections exceeding 1 step result in “skipping steps” to the next equilibrium point, which is 4 steps away.