- 20 Nov, 2024 5 commits
-
-
Lianmin Zheng authored
-
Ying Sheng authored
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
Ke Bao authored
-
- 19 Nov, 2024 7 commits
-
-
Lianmin Zheng authored
-
Yineng Zhang authored
-
HAI authored
-
Yineng Zhang authored
-
Alexander Waitz authored
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
- 18 Nov, 2024 15 commits
-
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
Jani Monoses authored
-
Yineng Zhang authored
-
yukavio authored
Co-authored-by:kavioyu <kavioyu@tencent.com>
-
Lianmin Zheng authored
Co-authored-by:
kavioyu <kavioyu@gmail.com>
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
Tanjiro authored
Co-authored-by:Tushar Goel <114812108+AI-Tushar@users.noreply.github.com>
-
DarkSharpness authored
-
Lianmin Zheng authored
-
Ke Bao authored
-
Lianmin Zheng authored
-
- 17 Nov, 2024 5 commits
-
-
Lianmin Zheng authored
-
Lianmin Zheng authored
-
Yineng Zhang authored
-
Lianmin Zheng authored
Co-authored-by:Haotian Liu <6631389+haotian-liu@users.noreply.github.com>
-
Lianmin Zheng authored
Fix illegal memory access in overlap mode & Use more fused triton kernels for building meta data (#2051)
-
- 16 Nov, 2024 5 commits
-
-
Ke Bao authored
-
Lianmin Zheng authored
-
HAI authored
-
Ke Wen authored
-
HAI authored
-
- 15 Nov, 2024 3 commits
-
-
Xiaoyu Zhang authored
-
Lianmin Zheng authored
-
Lianmin Zheng authored
[Fix] Adjust default chunked prefill size and cuda graph max bs according to GPU memory capacity (#2044)
-
