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 13, 19:30 UTC
[CI/Build] Enable some fixed tests in AMD CI (#28078)
Signed-off-by: zhewenli <zhewenli@meta.com>
Zhewen Li committed
Nov 4, 2025 at 19:15 UTC
878fd5a16f349d9518f4a0121d28b3cd0d48b869
1 file changed
+10
-10
.buildkite/test-amd.yaml
+10
-10
@@ -48,8 +48,8 @@ steps:
48
commands:
49
- bash standalone_tests/pytorch_nightly_dependency.sh
50
51
-- label: Async Engine, Inputs, Utils, Worker Test # 36min
52
- timeout_in_minutes: 50
51
+- label: Async Engine, Inputs, Utils, Worker Test # 10min
52
+ timeout_in_minutes: 15
53
mirror_hardwares: [amdexperimental, amdproduction]
54
agent_pool: mi325_1
55
# grade: Blocking
@@ -616,9 +616,9 @@ steps:
616
- uv pip install --system torchao==0.13.0
617
- VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization/ --ignore quantization/test_blackwell_moe.py
618
619
-- label: LM Eval Small Models # 53min
620
- timeout_in_minutes: 75
621
- mirror_hardwares: [amdexperimental]
619
+- label: LM Eval Small Models # 15min
620
+ timeout_in_minutes: 20
621
+ mirror_hardwares: [amdexperimental, amdproduction]
622
agent_pool: mi325_1
623
# grade: Blocking
624
source_file_dependencies:
@@ -627,8 +627,8 @@ steps:
627
commands:
628
- pytest -s -v evals/gsm8k/test_gsm8k_correctness.py --config-list-file=configs/models-small.txt --tp-size=1
629
630
-- label: OpenAI API correctness # 22min
631
- timeout_in_minutes: 30
630
+- label: OpenAI API correctness # 10min
631
+ timeout_in_minutes: 15
632
mirror_hardwares: [amdexperimental, amdproduction]
633
agent_pool: mi325_1
634
# grade: Blocking
@@ -859,10 +859,10 @@ steps:
859
- pytest -v -s models/multimodal -m core_model --ignore models/multimodal/generation/test_whisper.py --ignore models/multimodal/processing
860
- 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
861
862
-- label: Multi-Modal Accuracy Eval (Small Models) # 50min
863
- mirror_hardwares: [amdexperimental]
862
+- label: Multi-Modal Accuracy Eval (Small Models) # 10min
863
+ mirror_hardwares: [amdexperimental, amdproduction]
864
agent_pool: mi325_1
865
- timeout_in_minutes: 70
865
+ timeout_in_minutes: 15
866
working_dir: "/vllm-workspace/.buildkite/lm-eval-harness"
867
source_file_dependencies:
868
- vllm/multimodal/