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 11, 20:00 UTC
Mark qwen2_5_vl as xfail (#30283)
Signed-off-by: Yanan Cao <gmagogsfm@gmail.com>
Yanan Cao committed
Dec 8, 2025 at 17:14 UTC
7b35011ad136748b3d18d069103feda9a239459f
1 file changed
+1
tests/compile/fullgraph/test_multimodal_compile.py
+1
@@ -17,6 +17,7 @@ def test_compile():
17
# forked needed to workaround https://github.com/vllm-project/vllm/issues/21073
18
@pytest.mark.forked
19
@pytest.mark.skipif(not current_platform.is_cuda(), reason="Skip if not cuda")
20
+@pytest.mark.xfail
21
def test_qwen2_5_vl_compilation(vllm_runner, monkeypatch):
22
"""Test that Qwen2.5-VL vision submodules are compiled.
23