[Fix] Fix gpt oss triton kernels and upgrade flashinfer back to 0.6.11.post1 (#25335)

Co-authored-by: sglang-bot <sglang-bot@users.noreply.github.com>
Co-authored-by: b8zhong <b8zhong@users.noreply.github.com>
Co-authored-by: mmangkad <mmangkad@users.noreply.github.com>
This commit is contained in:
Baizhou Zhang
2026-05-15 01:04:56 -07:00
committed by GitHub
co-authored by sglang-bot b8zhong mmangkad
parent 4adfc6cf7e
commit 0c19540550
13 changed files with 147 additions and 53 deletions
@@ -20,12 +20,7 @@ import torch
from sglang.test.ci.ci_register import register_cuda_ci
register_cuda_ci(
est_time=120,
stage="stage-b",
runner_config="1-gpu-large",
disabled="broken on main, see #24816",
)
register_cuda_ci(est_time=120, stage="stage-b", runner_config="1-gpu-large")
flashinfer_fused_moe = pytest.importorskip("flashinfer.fused_moe")