Three things:
- You do not need to use that many k-points to sample the 2D Brillouin zone in the self-consistent calculation of bulk MoTe2. So, you may consider reducing it from 41x41x1 to, e.g., 12x12x1 Gamma-centered k-point grid.
- I would suggest changing the pseudopotential/basis set to SG15-type/Medium basis set.
- In principle, you do not need to set the same ATK-DFT calculator to do the band structure analysis after geometry optimization.
If you make the changes mentioned, you must be able to do this calculation rather quickly, at least it worked for me, taking 6 minutes on my laptop.