DeepSeek/DeepGEMM
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepGEMM synced 2025-05-06 19:34:23 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Add two more optimization TODOs

Browse Source
This commit is contained in:
Chenggang Zhao 2025-04-27 17:51:11 +08:00
parent 33e0c3ce40
commit 86afd0c212
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@ -18,6 +18,8 @@ Despite its lightweight design, DeepGEMM's performance matches or exceeds expert
- [x] MoE scheduler with TMA multicast compatibility - [x] MoE scheduler with TMA multicast compatibility
- [x] Fix TMA multicast compatibility for indivisible shapes - [x] Fix TMA multicast compatibility for indivisible shapes
- [ ] Skip useless computation on M - [ ] Skip useless computation on M
- [ ] Share pipeline stages between scheduled blocks
- [ ] TMA store pipeline
- [ ] NVRTC as a faster compiler - [ ] NVRTC as a faster compiler
- [ ] Sanitizer for testing - [ ] Sanitizer for testing
- [ ] Weight gradient kernels for dense models - [ ] Weight gradient kernels for dense models