@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 1a59078c87
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

[CI/Build] Avoid duplicate ViT CG test introduced by accident (#45654)

Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>

Isotr0py committed Jun 17, 2026 at 20:23 UTC 1a59078c873fb6dd14f23f17307a24977cdd9228
1 file changed +8 -2
.buildkite/test_areas/models_multimodal.yaml
+8 -2
@@ -30,7 +30,6 @@ steps:
30 - pip install git+https://github.com/TIGER-AI-Lab/Mantis.git
31 - pytest -v -s models/multimodal/generation/test_common.py -m core_model -k "qwen3 or gemma"
32 - pytest -v -s models/multimodal/generation/test_qwen2_5_vl.py -m core_model
33 - - pytest -v -s models/multimodal/generation/test_vit_cudagraph.py -m core_model
33 mirror:
34 amd:
35 device: mi325_1
@@ -63,9 +62,16 @@ steps:
62 - tests/models/multimodal
63 commands:
64 - pip install git+https://github.com/TIGER-AI-Lab/Mantis.git
66 - - pytest -v -s models/multimodal -m core_model --ignore models/multimodal/generation/test_common.py --ignore models/multimodal/generation/test_ultravox.py --ignore models/multimodal/generation/test_qwen2_5_vl.py --ignore models/multimodal/generation/test_qwen2_vl.py --ignore models/multimodal/generation/test_whisper.py --ignore models/multimodal/generation/test_memory_leak.py --ignore models/multimodal/processing
65 + - pytest -v -s models/multimodal -m core_model --ignore models/multimodal/generation/test_common.py --ignore models/multimodal/generation/test_ultravox.py --ignore models/multimodal/generation/test_qwen2_5_vl.py --ignore models/multimodal/generation/test_qwen2_vl.py --ignore models/multimodal/generation/test_whisper.py --ignore models/multimodal/generation/test_memory_leak.py --ignore models/multimodal/generation/test_vit_cudagraph.py --ignore models/multimodal/processing
66 + - pytest -v -s models/multimodal/generation/test_vit_cudagraph.py -m core_model
67 - pytest models/multimodal/generation/test_memory_leak.py -m core_model
68 - cd .. && VLLM_WORKER_MULTIPROC_METHOD=spawn pytest -v -s tests/models/multimodal/generation/test_whisper.py -m core_model # Otherwise, mp_method="spawn" doesn't work
69 + mirror:
70 + amd:
71 + soft_fail: true
72 + device: mi325_1
73 + depends_on:
74 + - image-build-amd
75
76 - label: Multi-Modal Processor (CPU)
77 key: multi-modal-processor-cpu