I did measurement from Reset to main (not loop!) some time ago and measured 2.12ms for M0+ and an additional 0.36ms for the M4.
Keep in mind that the startup also configures the clocks and waits for them to settle, as well as initializes the RAM, so there might be some variation in your project. I was using the WCO and IMO with FLL at 50MHz.
For reaching the main loop you have to consider your peripheral init, every peripheral initialization will add time (some more, some less), so this is a highly project specific value.
May I ask what application this is meant for?
thank you very much for the reply and sorry for the delay.
It is a locker application.