A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 9, 18:15 UTC
[CPU Backend][BugFix] Fix failing CPU MoE test (#32876)
Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
Fadi Arafeh committed
Jan 23, 2026 at 12:06 UTC
aac0b817fad9947608157529e68aa867ff76c804
1 file changed
+1
tests/kernels/moe/test_moe.py
+1
@@ -1311,6 +1311,7 @@ def test_moe_sum(m: int, topk: int, k: int, dtype: torch.dtype):
1311
opcheck(torch.ops._moe_C.moe_sum, (input, actual))
1312
1313
1314
+@pytest.mark.usefixtures("default_vllm_config")
1315
@pytest.mark.parametrize("m", [1, 33])
1316
@pytest.mark.parametrize("n,k", [(128, 128)])
1317
@pytest.mark.parametrize("e", [8])