Releases: flash-algo/flash-sparse-attention
Releases · flash-algo/flash-sparse-attention
v1.1.3
What's Changed
- Add support for targeted GPU architecture builds by @LoserCheems in #171
Full Changelog: v1.1.0...v1.1.3
v1.1.2
What's Changed
- Add support for targeted GPU architecture builds by @LoserCheems in #171
Full Changelog: v1.1.0...v1.1.2
v1.1.1
What's Changed
- Add support for targeted GPU architecture builds by @LoserCheems in #171
Full Changelog: v1.1.0...v1.1.1
v1.1.0
What's Changed
- [BUG FIX] Prevent mask/bias materialization; avoid OOB for irregular seqlen by @LoserCheems in #168
- [FEATURE SUPPORT] Optional mask/bias (3D & 4D) by @LoserCheems in #170
Full Changelog: v1.0.7...v1.1.0
v1.0.7
What's Changed
- Update issue templates and add auto-assignment for PRs by @LoserCheems in #158
- Add manual PyPI publishing workflow by @LoserCheems in #159
- Update installation requirements and streamline process by @LoserCheems in #160
- Add tensor operation utilities and performance enhancements by @LoserCheems in #165
- [FEATURE SUPPORT] Convert attention mask storage from float to bool by @LoserCheems in #166
- [FEATURE SUPPORT] Flexible head dims for mask/bias with in-kernel conversion path by @LoserCheems in #167
Full Changelog: v1.0.6...v1.0.7
v1.0.6
What's Changed
- Add wheel URL generation and caching for pre-built wheels in setup.py by @LoserCheems in #156
- Update PyTorch version matrix for Python 3.13 support by @LoserCheems in #157
Full Changelog: v1.0.5...v1.0.6
v1.0.5
Full Changelog: v1.0.4...v1.0.5
v1.0.4
What's Changed
- Add GitHub Actions workflows for building and publishing wheels by @LoserCheems in #155
Full Changelog: v1.0.3...v1.0.4
v1.0.3
What's Changed
- Add forward and backward performance tables to README and README_zh by @LoserCheems in #152
- Simplify backend imports and initialization by @LoserCheems in #153
- Update-docs by @LoserCheems in #154
Full Changelog: v1.0.2...v1.0.3
v1.0.2
What's Changed
- add correct arch info for hopper platform by @yiakwy-xpu-ml-framework-team in #149
New Contributors
- @yiakwy-xpu-ml-framework-team made their first contribution in #149
Full Changelog: v1.0.1...v1.0.2