deepseek-ai / deepseek-ai/FlashMLA
RuntimeError: Sparse BF16 MLA is not supported on SM90
Open
- Dominant language
- C++
- Stars
- 12.9k
- Forks
- 1.2k
- Avg merge
- 4h 20m
- Merged PRs (30d)
- 2
Description
请教为啥 FlashMLA 对数据类型的支持不完善,只支持 fp8,这对低于 sm90 的架构无法使用呀,请问有支持计划嘛?
建议可以支持一下 hopper + bf16,顺便可以对比下 fp8 带来的性能提升?
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.