@User @User @User Regarding RF. I want to remark that this is specific to Sky130 and for example for TSMC it is different. There one can have a more accurate model for RF of the same transistor. This is often the case because first the logic/MS version of a process is released with first set of models. And later when the RF version of the process with more RF primitives is released one also get a more accurate model of the 'logic' transistor. Switching between models in then often done by drawing a marker layer over the transistor but on the chip it will be the same transistor wether or not it is covered with that marker. The model improvement exist in better modeling of internal parasitics to get better matching at high frequencies.