Ridge: $\hat{\boldsymbol{\beta}}\to\mathbf{0}$ but never exactly zero. Lasso: can shrink coefficients to exactly zero (variable selection)

Both ridge and Lasso produce exactly sparse solutions

Ridge produces sparse solutions; Lasso does not

Both converge to the OLS estimate