[CI] fix test_qwen3_moe_external_launcher_ep_tp2 (#6951)

### What this PR does / why we need it?
fix test_qwen3_moe_external_launcher_ep_tp2 by
wait_until_npu_memory_free

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.16.0
- vLLM main:
https://github.com/vllm-project/vllm/commit/15d76f74e2fdb12a95ea00f0ca283acf6219a2b7

Signed-off-by: hfadzxy <starmoon_zhang@163.com>
This commit is contained in:
zhangxinyuehfad authored and GitHub committed 2026-03-05 16:43:45 +08:00
1 parent 1f2a083597
commit a6745b8577
1 file changed
+2
@@ -79,6 +79,7 @@ def test_qwen3_external_launcher(model):
@pytest.mark.parametrize("model", MOE_MODELS)
@wait_until_npu_memory_free()
def test_qwen3_moe_external_launcher_ep_tp2(model):
script = Path(
__file__
@@ -208,6 +209,7 @@ def test_qwen3_external_launcher_with_sleepmode_level2():
reason="This test is only for Ascend910B devices.",
)
@pytest.mark.parametrize("model", MODELS)
@wait_until_npu_memory_free()
@patch.dict(os.environ, {
"VLLM_ASCEND_ENABLE_MATMUL_ALLREDUCE": "1",
"HCCL_BUFFSIZE": "500"