hybrid should be faster (depending on mask and masklength) due to the fact, that hashcat uses 2 loops for that and the modloop/maskloop is executed directly on the gpu working as amplifier for each given word
for more details see
https://hashcat.net/wiki/doku.php?id=fre...full_speed
for more details see
https://hashcat.net/wiki/doku.php?id=fre...full_speed