Commit Graph

3 Commits

Author SHA1 Message Date
pyp_l40
7121981bb4 change default inference from top-p to top-k sampling, massive performance gain 2025-03-15 18:16:27 -05:00
Forkoz
1e79d9032e Empty cuda cache between inferences 2024-04-06 00:05:06 +00:00
jason-on-salt-a40
6760f29bd0 init 2024-03-21 11:02:20 -07:00