[Scheduler][Overlap] Skip decode phase for max_new_tokens=1 requests #67830
pr-test.yml
on: pull_request
check-changes
9s
Matrix: Build Wheel
Matrix: Build Wheel Arm
Matrix: multimodal-gen-test-1-gpu
Matrix: multimodal-gen-test-2-gpu
sgl-kernel-unit-test
sgl-kernel-mla-test
sgl-kernel-benchmark-test
stage-a-test-1
Matrix: stage-b-test-large-1-gpu
Matrix: stage-b-test-large-2-gpu
Matrix: stage-b-test-small-1-gpu
Matrix: stage-c-test-4-gpu-b200
Matrix: stage-c-test-4-gpu-h100
Matrix: stage-c-test-8-gpu-h20
Matrix: stage-c-test-8-gpu-h200
stage-c-test-deepep-4-gpu
0s
stage-c-test-deepep-8-gpu-h200
0s
stage-c-test-large-4-gpu-b200
0s
pr-test-finish
2s
Annotations
2 errors
|
call-gate / pr-gate
Process completed with exit code 1.
|
|
pr-test-finish
Process completed with exit code 1.
|