I believe the effect is so slight that you'd need galaxy size masses and really large astronomical distances to see something.
For the naked eye, perhaps, but the
Gaia mission, which is capable of measuring positions of stars to 7 millionths of an arcsecond, will be taking account of relativistic light effects from Sol (and perhaps even Earth).
Light paths around a black hole depend very much on how close to the event horizon the path dips:
One cool thing about gravitational lensing is that one can work backwards from the observed bending to a mass distribution. For example, from the observed lensing of a background galaxies by galaxy cluster CL0024+17:
it was possible to calculate a mass distribution that could account for it:
This mass distribution is hugely important for providing constraints to theories of
dark matter (and perhaps also for its theoretical competitor
modified Newtonian mechanics)