Skip to content

[TRITON] Add attention sink support to Triton MHA kernels (#1576) #1

[TRITON] Add attention sink support to Triton MHA kernels (#1576)

[TRITON] Add attention sink support to Triton MHA kernels (#1576) #1

Triggered via push December 11, 2025 11:05
Status Cancelled
Total duration 1d 0h 0m 53s
Artifacts

aiter-test.yaml

on: push
check-signal
38s
check-signal
define-runners
4s
define-runners
Matrix: multi-gpu
Matrix: standard
Fit to window
Zoom out
Zoom in

Annotations

2 errors
multi-gpu (aiter-mi355-8gpu)
The job has exceeded the maximum execution time while awaiting a runner for 24h0m0s
standard (aiter-mi355-1gpu)
The job has exceeded the maximum execution time while awaiting a runner for 24h0m0s