Magnitude‐rupture scaling relations describe how the length, width, and area of fault rupture vary with earthquake magnitude. These parameters are required in seismic hazard models to fit the models’ earthquakes onto faults and to define the site‐rupture distances needed in ground‐motion prediction equations. We collected the magnitude and rupture parameters of 91 earthquakes in Mainland China and nearby regions to study magnitude‐rupture scaling relations. We find no systematic deviations for the subsurface rupture length (RLD) obtained from different methods versus earthquake magnitude. We performed regressions of RLD versus magnitude and versus rupture width using general orthogonal regression. Then, we derived the relations between rupture area and magnitude. Our relations are not statistically different from the results derived by others using global datasets, if the parameters of the five pre‐1900 great earthquakes in eastern China are not used. However, if the five earthquakes are used, the magnitude‐rupture length scaling relation for large strike‐slip earthquakes in eastern China gives shorter rupture lengths than earthquakes in western China and other plate boundary regions in the world.