Skip to content

Actions: fla-org/flash-linear-attention

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
2,500+ workflow runs
2,500+ workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
[Model] Add CAT (Compress and Attend Transformer) nvidia-h100-ci #4602: Pull request #1017 synchronize by bicycleman15
Queued bicycleman15:cat_pr
bicycleman15:cat_pr
Queued
[Model] Add CAT (Compress and Attend Transformer) lint #5493: Pull request #1017 synchronize by bicycleman15
30s bicycleman15:cat_pr
bicycleman15:cat_pr
30s
[Model] Add CAT (Compress and Attend Transformer) ascend-a2-benchmark-ci #601: Pull request #1017 synchronize by bicycleman15
6s bicycleman15:cat_pr
bicycleman15:cat_pr
6s
[Model] Add CAT (Compress and Attend Transformer) ascend-a2-ci #960: Pull request #1017 synchronize by bicycleman15
6s bicycleman15:cat_pr
bicycleman15:cat_pr
6s
[Model] Add CAT (Compress and Attend Transformer) check-header #1657: Pull request #1017 synchronize by bicycleman15
12s bicycleman15:cat_pr
bicycleman15:cat_pr
12s
[Model] Add CAT (Compress and Attend Transformer) check-pr-title #990: Pull request #1017 synchronize by bicycleman15
5s bicycleman15:cat_pr
bicycleman15:cat_pr
5s
[Model] Add CAT (Compress and Attend Transformer) check-no-oj #502: Pull request #1017 synchronize by bicycleman15
7s bicycleman15:cat_pr
bicycleman15:cat_pr
7s
[Ops] Add Triton-Ascend compatibility for GDN-2 nvidia-h100-ci #4601: Pull request #1213 submitted by zheliuyu
Pending
Pending
Post benchmark comment Post benchmark comment #1869: completed by zhiyuan1i
9s
9s
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16 mmad, and fused a2/dg nvidia-h100-ci #4600: Pull request #1209 closed by zhiyuan1i
8s sunyi0505:prepare_wy_repr_bwd_kv_npu
sunyi0505:prepare_wy_repr_bwd_kv_npu
8s
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16… nvidia-h100-ci #4599: Commit 8e84ed4 pushed by zhiyuan1i
Queued main
main
Queued
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16… ascend-a2-ci #959: Commit 8e84ed4 pushed by zhiyuan1i
1h 57m 31s main
main
1h 57m 31s
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16… lint #5492: Commit 8e84ed4 pushed by zhiyuan1i
29s main
main
29s
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16… ascend-a2-benchmark-ci #600: Commit 8e84ed4 pushed by zhiyuan1i
1h 13m 52s main
main
1h 13m 52s
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16… check-header #1656: Commit 8e84ed4 pushed by zhiyuan1i
8s main
main
8s
Code Quality: Push on main CodeQL #49: by zhiyuan1i
2m 51s main
main
2m 51s
Post benchmark comment Post benchmark comment #1868: completed by zhiyuan1i
1s
1s
Post benchmark comment Post benchmark comment #1867: completed by zhiyuan1i
9s
9s
[Perf] Replace split Ascend KDA bwd_intra kernels with a single persistent kernel nvidia-h100-ci #4598: Pull request #1210 submitted by zhiyuan1i
Pending
Pending
[Perf] Optimize Ascend prepare_wy_repr_bwd with larger kv tiles, bf16 mmad, and fused a2/dg nvidia-h100-ci #4597: Pull request #1209 submitted by zhiyuan1i
6s
6s
[Perf] Replace split Ascend KDA bwd_intra kernels with a single persistent kernel nvidia-h100-ci #4596: Pull request #1210 submitted by zhiyuan1i
1m 40s
1m 40s
Post benchmark comment Post benchmark comment #1866: completed by zhiyuan1i
1s
1s
[Models] Add MomentumDeltaNet implementation from arXiv:2605.05838 (ICML2026) nvidia-h100-ci #4595: Pull request #1208 submitted by zhiyuan1i
8s
8s
Post benchmark comment Post benchmark comment #1865: completed by hazelduan
8s
8s
[Ops] Add Triton-Ascend compatibility for GDN-2 nvidia-h100-ci #4594: Pull request #1213 synchronize by hazelduan
Queued hazelduan:gdn2-ascend-compat
hazelduan:gdn2-ascend-compat
Queued