Commit Graph

5 Commits (1b4e1ff0ef60c27fe81f7189909af2ec4eef3a76)

Author SHA1 Message Date
Seunghoon Lee 18f58e13f8
Update fwd_prefill.py 2025-05-09 13:36:01 +09:00
AS 5e561560ea
Update fwd_prefill.py
Remove cudagraph from triton autotune. Buggy under zluda
2025-05-08 19:32:51 -07:00
Seunghoon Lee 2a4eb86d10
zluda 3.9.4 & update flash attention 2 2025-05-07 00:39:40 +09:00
Seunghoon Lee 854b49f7b4
remove redundant parts zluda fa 2025-04-28 13:52:24 +09:00
Seunghoon Lee ab9e87d848
zluda flash attention 2 via triton 2025-03-22 23:10:59 +09:00