kernels-community
/

paged-attention

Model card Files Files and versions Community

paged-attention / build.toml

Commit History

Use default CUDA capabilities

0f86240

danieldk HF Staff commited on 18 days ago

Sync capabilities with upstream

3f98f45

danieldk HF Staff commited on 21 days ago

Update for build.toml changes

6df83c0

danieldk HF Staff commited on Feb 27

Rename to paged-attention

3dcba92

danieldk HF Staff commited on Feb 5

Port vLLM attention kernels

132e594

danieldk HF Staff commited on Jan 29