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 6, 05:45 UTC
[MRV2] Enable PP CUDA graph test (#37830)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Woosuk Kwon committed
Mar 22, 2026 at 16:30 UTC
43877a620bf629d3625c870ef787e590101e0518
1 file changed
+2
-3
.buildkite/test_areas/model_runner_v2.yaml
+2
-3
@@ -87,13 +87,12 @@ steps:
87
- vllm/v1/worker/gpu/
88
- vllm/v1/worker/gpu_worker.py
89
- tests/distributed/test_pipeline_parallel.py
90
- #- tests/distributed/test_pp_cudagraph.py
90
+ - tests/distributed/test_pp_cudagraph.py
91
commands:
92
- set -x
93
- export VLLM_USE_V2_MODEL_RUNNER=1
94
- pytest -v -s distributed/test_pipeline_parallel.py -k "not ray and not Jamba"
95
- # TODO: Uncomment once https://github.com/vllm-project/vllm/pull/35162 is merged.
96
- #- pytest -v -s distributed/test_pp_cudagraph.py -k "not ray"
95
+ - pytest -v -s distributed/test_pp_cudagraph.py -k "not ray"
96
97
- label: Model Runner V2 Spec Decode
98
timeout_in_minutes: 30