Reliable detection, post-processing and fitting of MM is crucial for quantifying short-TE 1H-MR brain spectra. In general, an in-vivo acquired full-MM spectrum is included in the basis-set together with a relatively free spline baseline (i.e.LCModel). Availability of high magnetic fields (better resolved in-vivo MM) lead to a need for more sophisticated approaches such as MM parametrization. Furthermore, effect of the stiffness of fitted baseline on the resulting metabolite concentrations gained a lot of interest. We compared parametrized and full-MM basis-sets, with varying DKNTMN to assess the resulting changes in metabolite concentrations (in-vivo rat brain 1H-MRS and Monte-Carlo simulations at 9.4T).