GROUP BY vs DISTINCT

Which is more efficient GROUP BY or DISTINCT to find duplicates in Teradata?

With more duplicates GROUP BY is more efficient while if we have fewer duplicates the DISTINCT is efficient.