Commits · 5474be529912fba324fe944f5e32fa1879a09d51 · wqshmzh / ktransformers

25 Feb, 2025 1 commit
- support absorb for prefill long context · f4c198bd
  Atream authored Feb 25, 2025
  
  f4c198bd
23 Feb, 2025 1 commit
- tmp · 95d937c5
  DDong Jianwei authored Feb 23, 2025
  
  95d937c5
19 Feb, 2025 1 commit
- clean PR code and disable flashinfer · a5295183
  Atream authored Feb 19, 2025
  
  a5295183
18 Feb, 2025 1 commit
- fix: adapt prefix cache in `forward_linux_flashinfer` · 2ffb43f9
  ceerrep authored Feb 18, 2025
  
  2ffb43f9
17 Feb, 2025 1 commit
- fix precision bug imported by position_ids in 0.2.0 · 038bc308
  Atream authored Feb 17, 2025
  
  038bc308
16 Feb, 2025 2 commits
- fix: use flash_attn for faster prefill · 5ac26608
  ceerrep authored Feb 17, 2025
  
  5ac26608
- Mock triton mla due to precision issue · ff6b265e
  Azure authored Feb 16, 2025
  
  ff6b265e
15 Feb, 2025 1 commit
- Update attention.py · 92399283
  Atream authored Feb 15, 2025
  
  92399283
14 Feb, 2025 1 commit
- linux support triton MLA kernel · 1084d4e4
  Atream authored Feb 14, 2025
  
  1084d4e4
13 Feb, 2025 1 commit
- init support for MLA using Attention kernel · bb35dc5b
  Atream authored Feb 13, 2025
  
  bb35dc5b
04 Feb, 2025 1 commit
- done support deepseekv3 · 907251c7
  Azure authored Feb 04, 2025
  
  907251c7
01 Feb, 2025 1 commit
- update rope calculation; update modeling.py; update gate for moe · f873558a
  Azure authored Feb 01, 2025
  
  f873558a
31 Jan, 2025 1 commit
- support deepseekv3; runable but have precition problem · 476b1d8d
  Azure authored Jan 31, 2025
  
  476b1d8d
02 Sep, 2024 1 commit
- fix qlen > 1000 mask is none error · c55de02f
  Azure authored Sep 02, 2024
  
  c55de02f
28 Aug, 2024 1 commit
- [feature] release 0.1.3 · 4d1d561d
  chenxl authored Aug 28, 2024
  
  4d1d561d
15 Aug, 2024 1 commit
- [fix] format classes and files name · 67043b4b
  TangJingqi authored Aug 15, 2024
  
  67043b4b
27 Jul, 2024 1 commit
- Initial commit · 18c42e67
  chenxl authored Jul 27, 2024
  
  18c42e67