Instructions to use kernels-community/paged-attention with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Kernels
How to use kernels-community/paged-attention with Kernels:
# !pip install kernels from kernels import get_kernel # a version (or an explicit revision) is required; see the "Files and versions" tab for the available ones kernel = get_kernel("kernels-community/paged-attention", version=1) - Notebooks
- Google Colab
- Kaggle
Commit History
Update tag 4ca355d
Build (aarch64-linux) 5e584b3
Build (x86_64-linux) b4c51e9
Update flake 0041e3f
Build (x86_64-linux Torch 2.8) d17b8f3
Build (aarch64-linux Torch 2.8) f207cd2
Build (x86_64-linux) 8b32f11
Build (aarch64-linux) e6ce28c
Enable ROCm build 6677800
Sync CUDA paged-attention with upstream 4c6b316
Improve tests for mps 4dbd9c0
Update readme 16fc7e4
Add metal paged attention ed30f9d
feat: add tag for hfjob build a0903d3 verified
Build 1e0a970
Build (AArch64) 20990f8
Update flake inputs daf6221
Use default CUDA capabilities 0f86240
Fix flake input bebc17e
Build (aarch64) a9bb8f7
Build dde9676
Sync capabilities with upstream 3f98f45
Update flake cea0337
feat: update to include rev in kernel for reproducible symbols 9164b48
drbh commited on