deepseek-ai / deepseek-ai/DeepGEMM
Will swap ab tech improve small m (decode phase) performance?
Open
- Dominant language
- Cuda
- Stars
- 7.8k
- Forks
- 1.3k
- Avg merge
- 3d 7h
- Merged PRs (30d)
- 3
Description
As title described.
Contributor guide
No contributing guide indexed for this repository
Research direction
The issue names no files, tests, or entry points and only asks whether swapping the AB technology improves decode-phase performance. Start by locating the decode-phase kernels and their existing benchmarks, then compare the alternatives and document a measured result and conclusion.
Written by the indexing model from the issue text.
Assessment
- Domain
- hpc, performance
- Issue type
- Refactor
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100