From 4e4bbc42a6f4d511ec6c3542bbb439afbe563892 Mon Sep 17 00:00:00 2001 From: TJian Date: Wed, 1 Apr 2026 19:37:34 +0800 Subject: [PATCH] [CI] Tune GPU resources for test (#2401) Signed-off-by: tjtanaa --- .buildkite/test-merge.yml | 2 +- .buildkite/test-ready.yml | 4 ++-- tests/e2e/online_serving/test_mimo_audio.py | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/.buildkite/test-merge.yml b/.buildkite/test-merge.yml index 7bee19319..5ee936337 100644 --- a/.buildkite/test-merge.yml +++ b/.buildkite/test-merge.yml @@ -174,7 +174,7 @@ steps: pytest -s -v tests/engine/test_async_omni_engine_abort.py ' agents: - queue: "gpu_4_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU + queue: "gpu_1_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU plugins: - docker#v5.2.0: image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT diff --git a/.buildkite/test-ready.yml b/.buildkite/test-ready.yml index 89839a2d1..91ea92a5c 100644 --- a/.buildkite/test-ready.yml +++ b/.buildkite/test-ready.yml @@ -180,7 +180,7 @@ steps: pytest -s -v tests/engine/test_async_omni_engine_abort.py ' agents: - queue: "gpu_4_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU + queue: "gpu_1_queue" # g6.12xlarge instance on AWS, has 4 L4 GPU plugins: - docker#v5.2.0: image: public.ecr.aws/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT @@ -271,7 +271,7 @@ steps: - image: 936637512419.dkr.ecr.us-west-2.amazonaws.com/vllm-ci-pull-through-cache/q9t5s3a7/vllm-ci-test-repo:$BUILDKITE_COMMIT resources: limits: - nvidia.com/gpu: 2 + nvidia.com/gpu: 1 volumeMounts: - name: devshm mountPath: /dev/shm diff --git a/tests/e2e/online_serving/test_mimo_audio.py b/tests/e2e/online_serving/test_mimo_audio.py index 639c46a65..2fb63c1e4 100644 --- a/tests/e2e/online_serving/test_mimo_audio.py +++ b/tests/e2e/online_serving/test_mimo_audio.py @@ -95,7 +95,7 @@ def get_max_batch_size(size_type="few"): @pytest.mark.advanced_model @pytest.mark.core_model @pytest.mark.omni -@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards=2) +@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards=1) @pytest.mark.parametrize("omni_server", test_params, indirect=True) def test_audio_to_text_audio_001(omni_server, openai_client) -> None: """ @@ -128,7 +128,7 @@ def test_audio_to_text_audio_001(omni_server, openai_client) -> None: @pytest.mark.advanced_model @pytest.mark.omni -@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards=2) +@hardware_test(res={"cuda": "L4", "rocm": "MI325"}, num_cards=1) @pytest.mark.parametrize("omni_server", test_params, indirect=True) def test_text_to_text_001(omni_server, openai_client) -> None: """