From ec1d2b5c04586ae2088dd82b0a6cbafae007a241 Mon Sep 17 00:00:00 2001 From: Yizhou <136800916+yiz-liu@users.noreply.github.com> Date: Tue, 21 Oct 2025 17:16:15 +0800 Subject: [PATCH] [Test] Temporarily skip flaky ACL graph test (#3577) ### What this PR does / why we need it? Disables `FULL_DECODE_ONLY` end-to-end test that fails intermittently. This prevents CI blockages while the root cause of the flakiness is investigated. ### Does this PR introduce _any_ user-facing change? None. ### How was this patch tested? None needed. - vLLM version: v0.11.0rc3 - vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0 Signed-off-by: Yizhou Liu --- tests/e2e/singlecard/test_aclgraph.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/e2e/singlecard/test_aclgraph.py b/tests/e2e/singlecard/test_aclgraph.py index 570dbbc..efa6cb3 100644 --- a/tests/e2e/singlecard/test_aclgraph.py +++ b/tests/e2e/singlecard/test_aclgraph.py @@ -100,6 +100,8 @@ def test_models_with_aclgraph( ) +@pytest.mark.skip("Skipping this test for now, " + "it fails intermittently and needs investigation.") @pytest.mark.parametrize("model", MODELS) @pytest.mark.parametrize("max_tokens", [5]) def test_models_with_aclgraph_full_decode_only(