Seunghoon Lee
|
49965dfda8
|
get_hip_arch_name -> get_hip_agent, use amdhip64_7.dll served within rocm package
|
2026-01-03 21:00:36 +09:00 |
|
vladmandic
|
b9c18452f2
|
unify hip get arch name
Signed-off-by: vladmandic <mandic00@live.com>
|
2026-01-03 08:22:19 +01:00 |
|
vladmandic
|
f6289206d8
|
startup sequence optimizations
Signed-off-by: vladmandic <mandic00@live.com>
|
2026-01-03 07:54:05 +01:00 |
|
vladmandic
|
a78a99f0ff
|
use balanced for auto-detected 4gb gpu
Signed-off-by: vladmandic <mandic00@live.com>
|
2025-12-22 21:03:46 +01:00 |
|
Disty0
|
f4ee9c7052
|
Add Flex attention
|
2025-11-09 00:14:38 +03:00 |
|
Disty0
|
2bbbb684cc
|
Rename CK Flash attention to just Flash attention
|
2025-11-08 23:24:40 +03:00 |
|
Disty0
|
a93715e0da
|
Don't expose AMD Triton Flash Atten for non AMD
|
2025-11-08 23:20:55 +03:00 |
|
Disty0
|
93797dff8e
|
ROCm enable Dynamic atten only for RDNA2 and older GPUs
|
2025-11-08 22:53:26 +03:00 |
|
Vladimir Mandic
|
56026c4e61
|
refactor attention handling
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-11-08 10:55:41 -05:00 |
|
Disty0
|
c51552af90
|
Enable triton flash atten option for rocm linux too
|
2025-09-03 16:26:00 +03:00 |
|
Disty0
|
266c9c0d3d
|
Move Zluda Triton flash atten hijack to Triton Flash attention option
|
2025-09-03 16:16:41 +03:00 |
|
Vladimir Mandic
|
9e6928410f
|
offloading improvements
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-08-25 10:16:38 -04:00 |
|
Disty0
|
6597a9111f
|
Fix offload detection with CPU only envs
|
2025-08-16 16:12:43 +03:00 |
|
Vladimir Mandic
|
b120778c4f
|
lint
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-07-31 17:33:44 -04:00 |
|
Vladimir Mandic
|
fa44521ea3
|
offload-never and offload-always per-module and new highvram profile
Signed-off-by: Vladimir Mandic <mandic00@live.com>
|
2025-07-31 11:40:24 -04:00 |
|