Skip to content

Actions: flash-algo/flash-sparse-attention

Actions

Auto assign reviewers and assignees

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
auto-assign.yml
41 workflow runs
41 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Refactor attention mask and bias handling for efficiency Auto assign reviewers and assignees #16: Pull request #177 opened by LoserCheems
15h 18m 31s
15h 18m 31s
Bump version to 1.1.8 Auto assign reviewers and assignees #15: Pull request #176 opened by LoserCheems
51s
51s
Increase GitHub Actions build timeout to 6 hours Auto assign reviewers and assignees #14: Pull request #175 opened by LoserCheems
2m 37s
2m 37s
Remove CUDA architecture '120' for compatibility Auto assign reviewers and assignees #13: Pull request #174 opened by LoserCheems
10s
10s
Expand build matrix for ARM64 and additional CUDA architectures Auto assign reviewers and assignees #12: Pull request #173 opened by LoserCheems
1m 12s
1m 12s
Refine build matrix and CUDA architecture specifications Auto assign reviewers and assignees #11: Pull request #172 opened by LoserCheems
13s
13s
Add support for targeted GPU architecture builds Auto assign reviewers and assignees #10: Pull request #171 opened by LoserCheems
4m 51s
4m 51s
[FEATURE SUPPORT] Optional mask/bias (3D & 4D) Auto assign reviewers and assignees #9: Pull request #170 opened by LoserCheems
13s
13s
[BUG FIX] Prevent mask/bias materialization; avoid OOB for irregular seqlen Auto assign reviewers and assignees #8: Pull request #168 opened by LoserCheems
10s
10s
[FEATURE SUPPORT] Flexible head dims for mask/bias with in-kernel conversion path Auto assign reviewers and assignees #7: Pull request #167 opened by LoserCheems
11s
11s
[FEATURE SUPPORT] Convert attention mask storage from float to bool Auto assign reviewers and assignees #6: Pull request #166 opened by LoserCheems
11s
11s
Add tensor operation utilities and performance enhancements Auto assign reviewers and assignees #5: Pull request #165 opened by LoserCheems
13s
13s
Implement Unified Sparse Mask Strategy with Block-Level Skipping Auto assign reviewers and assignees #4: Pull request #164 opened by Copilot AI
Action required
Action required
Add optional mask & bias inputs with adaptive computation skipping Auto assign reviewers and assignees #3: Pull request #162 opened by Copilot AI
Action required
Action required
Update installation requirements and streamline process Auto assign reviewers and assignees #2: Pull request #160 opened by LoserCheems
10s
10s
Add manual PyPI publishing workflow Auto assign reviewers and assignees #1: Pull request #159 opened by LoserCheems
11s
11s