minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[CI] Broaden stage-b-test-1-gpu-large runner pool to H100 + H200 (#24080)

Browse Source 
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Kangyan-Zhou
2026-04-29 12:18:10 -07:00
committed by GitHub
co-authored by Claude Opus 4.7
parent 903e46d848
commit d4040e7010
2 changed files with 5 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.github/workflows/pr-test.yml
+4 -1
View File
@@ -820,7 +820,10 @@ jobs:
((needs.check-changes.outputs.main_package == 'true') || (needs.check-changes.outputs.sgl_kernel == 'true')) ((needs.check-changes.outputs.main_package == 'true') || (needs.check-changes.outputs.sgl_kernel == 'true'))
) )
) )
runs-on: 1-gpu-h100 # `1-gpu-h100-h200` is a shared label advertised by both the 1-GPU H100 and
# 1-GPU H200 runner pools, so this job can land on either. Multimodal_gen
# jobs in pr-test-multimodal-gen.yml continue to pin to `1-gpu-h100`.
runs-on: 1-gpu-h100-h200
timeout-minutes: 240 timeout-minutes: 240
strategy: strategy:
fail-fast: false fail-fast: false
scripts/ci/utils/slash_command_handler.py
+1 -1
View File
@@ -479,7 +479,7 @@ CUDA_SUITE_TO_RUNNER = {
"stage-a-test-1-gpu-small": "1-gpu-5090", "stage-a-test-1-gpu-small": "1-gpu-5090",
"stage-a-test-cpu": "ubuntu-latest", "stage-a-test-cpu": "ubuntu-latest",
"stage-b-test-1-gpu-small": "1-gpu-5090", "stage-b-test-1-gpu-small": "1-gpu-5090",
"stage-b-test-1-gpu-large": "1-gpu-h100", "stage-b-test-1-gpu-large": "1-gpu-h100-h200",
"stage-b-test-2-gpu-large": "2-gpu-h100", "stage-b-test-2-gpu-large": "2-gpu-h100",
"stage-b-test-4-gpu-b200": "4-gpu-b200", "stage-b-test-4-gpu-b200": "4-gpu-b200",
"stage-c-test-4-gpu-h100": "4-gpu-h100", "stage-c-test-4-gpu-h100": "4-gpu-h100",