Revert "[CI][XPU] Disable model runner V2 for XPU quantization test for some partially pre-quantized models" (#56394)
Signed-off-by: Chaojun Zhang <chaojun.zhang@intel.com>
Chaojun Zhang committed
Sep 14, 2026 at 11:40 UTC
39545e475d3627287ff69c25465dc0bd405f67e1
1 file changed
-6
.buildkite/intel_jobs/test-intel.yaml
-6
@@ -148,16 +148,10 @@ steps:
148
source_file_dependencies:
149
- vllm/
150
- .buildkite/intel_jobs/test-intel.yaml
151
- # TODO: XPU fused MoE + model runner V2 currently hangs on some
152
- # pre-quantized models, e.g.
153
- # quantization/test_online.py::test_online_quantization_loads_real_weights[nm-testing/tinysmokeqwen3moe-W4A16-first-only-CTstable]
154
- # Remove the VLLM_USE_V2_MODEL_RUNNER=0 override below once the hang is
155
- # root-caused and fixed.
151
commands:
152
- >-
153
bash .buildkite/scripts/hardware_ci/run-intel-test.sh
154
'cd tests &&
160
- export VLLM_USE_V2_MODEL_RUNNER=0 &&
155
pytest -v -s quantization/test_auto_round.py &&
156
pytest -v -s quantization/test_online.py'
157
- label: "XPU compressed tensors FP8 test"