deepseek-ai / deepseek-ai/FlashMLA
Looking forward to additional configs in Flash Attenion 4
Open
- Dominant language
- C++
- Stars
- 12.9k
- Forks
- 1.2k
- Avg merge
- 4h 20m
- Merged PRs (30d)
- 2
Description
The Triton implementation of the [Flash Attention v2](https://tridao.me/publications/flash2/flash2.pdf) is currently a work in progress.
Would be nice to keep up with this implementation:
https://github.com/Dao-AILab/flash-attention
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by reading the repository's Triton implementation and comparing it with the linked Flash Attention repository. The issue names no files, tests, or specific configurations, so the required scope and completion criteria need to be clarified before implementation can begin.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100