While recent advances in computational protein design have led to an increased experimental success rate designing stable single-state proteins, reliably tuning the dynamics of multi-state proteins remains a challenge. Here, we describe a generalizable method for tuning multi-state conformational equilibria using Frame2seq, a structure-conditioned sequence design method. We demonstrate that our approach is consistently predictive of changes to the ratio of conformational populations resulting from experimentally characterized point mutations, without requiring multiple sequence alignments. Further, we use Frame2seq to identify mutation sites that will result in the most significant shifts to conformational switch equilibria as computationally validated with AlphaFold2. The residue-scale accuracy offered by our approach will advance the development and modulation of natural and de novo dynamic proteins.