zhan3916
|
d4d889febb
|
Update flashMLA implementation on MACA (#1)
|
2026-07-06 14:38:08 +08:00 |
zhan3916
|
30f7561775
|
Performance: Add new tile for small bs
Signed-off-by: zhan3916 <zhan3916@gmail.com>
|
2025-03-25 11:54:33 +08:00 |
zhan3916
|
2e747f0a42
|
Fix: fix large input length offset overflow
|
2025-03-19 17:29:06 +08:00 |
zhan3916
|
640115cb88
|
Support splitKV>1
1. Support splitKV>1
2. Refactor code structure
3. Fix some readme typo
Signed-off-by: zhan3916 <zhan3916@gmail.com>
|
2025-03-06 13:57:41 +08:00 |
Kevin Zhang
|
f1999a14b4
|
1. Support MTP>1
2. Add full dependency for build and run
|
2025-02-25 21:45:03 +08:00 |