In the (spontaneous) Raman process, incident light $\hbar \omega_1$ scatters and transfers some energy $\hbar \omega$ to a vibrational excitation of molecule or solid.
Typically this is a very rare process, and only happens to one of every $10^9$ photons or so. So even if a sample is irradiated with a ultrafast pump laser of high intensity, Raman processes tend to not be significant enough to move the system out of equilibrium. This is in comparison to something like direct absorption processes which easily bring the system out of equilibrium and often dominate the response.
But the discussion above is for spontaneous Raman processes and makes me wonder, what about stimulated Raman processes? In the case of stimulated Raman, two photons come in with $\omega_1-\omega_2=\omega$, causing vibrational excitations to be much more efficiently created.
So my questions are
In practical cases, what is the efficiency of Stimulated Raman processes? In other words, for a given number of pump photon $n_1$ and Stokes photon $n_2$, what are some ballpark numbers for the number of vibrational excitations created?
Can stimulated Raman processes be so strong as to move a system out of equilibrium in a pump-probe setup?