How to compare parametric and Cox PH models predictions without Brier score?

Trying to compare AFT and Cox PH predictions. Concordance only gives out how good the model is ranking the predictions. Cox PH has a partial likelihood which cant be compared to AFT. Looking for some evaluation technique for predictions of both the models.

You will probably have to try an evaluation metric outside of lifelines. The brier score is one such metric, and I would be happy to see that contributed to lifelines!