[Misc] Add and enable Triton kernel unit tests on XPU (#45694)

Signed-off-by: pmanczak <pmanczak@users.noreply.github.com>
Signed-off-by: pmanczak <pawel.manczak@intel.com>
Co-authored-by: pmanczak <pmanczak@users.noreply.github.com>
This commit is contained in:
pmanczak
2026-08-07 02:53:03 -07:00
committed by GitHub
co-authored by pmanczak
parent 5ec47f3e48
commit 6b5bec7bed
5 changed files with 58 additions and 13 deletions
@@ -13,7 +13,7 @@ import torch
from vllm.platforms import current_platform
from vllm.utils.torch_utils import set_random_seed
device = "cuda"
device = current_platform.device_type
triton_scaled_mm_module = importlib.import_module(
"vllm.model_executor.layers.quantization.compressed_tensors.triton_scaled_mm"