I received your models and could investigate them from my side.
I got result values similar to yours in Discovery Live with my Quadro P5000 graphics card as well.
Discovery Live is intended to show the trends of the results so that the designer can make qualitative design explorations but for high fidelity results I would recommend you go with Discovery AIM.
In this example shared by you, you will notice that the contour of the Stress and Displacement in Discovery Live and Discovery AIM are strikingly similar even though the maximum values of the Equivalent stress are not really comparable.
Now that being said, I'd like to bring your attention to another point in this case which could be a matter of concern.
The edge at which the max stress value is developed is a rather sharp edge which may not be a practical condition. In FEA, such stress values at such sharp edges could be the result of 'Stress Singularity'.
To check for this factor, I run a couple more iterations of the analysis with a more refined mesh size.
At mesh size 1.5mm the Max. Equivalent stress was 714MPa.
At mesh size 1mm the Max. Equivalent stress was 914MPa.
As the result seems to have large variations, I don't think it can be relied upon.
If you would like to persist with this model, I'd recommend you add a fillet at the sharp edges or take measures to remove the singularity issue.