deepseek-ai / deepseek-ai/FlashMLA
consult questions about lower kv_lora_rank support
Open
- Dominant language
- C++
- Stars
- 12.9k
- Forks
- 1.2k
- Avg merge
- 4h 20m
- Merged PRs (30d)
- 2
Description
I wonder if FlashMLA supports MLA with parameters other than d(head_dim_q)=576 ?
For example, if I can test FlashMLA with head_dim_qk = 256 + 64 and head_dim_vo = 256 with just a small code change?
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.