From 5c5434e2d89fe7008ec7a6046faa65a69ba568ca Mon Sep 17 00:00:00 2001 From: "Kevin H. Luu" Date: Fri, 24 Jul 2026 01:14:48 -0700 Subject: [PATCH] Remove Quantization test parallelism (#49693) Signed-off-by: khluu Co-authored-by: OpenAI Codex --- .buildkite/test_areas/quantization.yaml | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/.buildkite/test_areas/quantization.yaml b/.buildkite/test_areas/quantization.yaml index 16782f727f..0a79b71121 100644 --- a/.buildkite/test_areas/quantization.yaml +++ b/.buildkite/test_areas/quantization.yaml @@ -24,8 +24,7 @@ steps: - uv pip install --system conch-triton-kernels # The SM90-only checkpoint currently contains a removed weight_chan_scale # parameter. It was not exercised by the previous L4 job. - - VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization/ --ignore quantization/test_blackwell_moe.py -k 'not test_compressed_tensors_w4a8_fp8' --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT - parallelism: 8 + - VLLM_TEST_FORCE_LOAD_FORMAT=auto pytest -v -s quantization/ --ignore quantization/test_blackwell_moe.py -k 'not test_compressed_tensors_w4a8_fp8' - label: Quantized Fusions device: h200_35gb @@ -68,5 +67,4 @@ steps: - vllm/model_executor/layers/quantization - tests/models/quantization commands: - - pytest -v -s models/quantization --shard-id=$$BUILDKITE_PARALLEL_JOB --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT - parallelism: 3 + - pytest -v -s models/quantization