SM-2 adjusts card intervals based on how you rate each review (Again/Hard/Good/Easy). It works well across the board and has 40+ years of evidence behind it.
FSRS uses your actual review history to fit a personal forgetting curve, which produces slightly more efficient scheduling — typically 10–20% fewer reviews for the same retention. The catch: it needs ~1,000 review data points to train, so brand-new users see no benefit.
QuickCards uses SM-2 today with FSRS support on the roadmap for Pro users. Most students will not see a major real-world difference either way.