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 11, 23:45 UTC
[CI/Build] Update batch invariant test trigger (#30080)
Signed-off-by: zhewenli <zhewenli@meta.com>
Zhewen Li committed
Dec 4, 2025 at 16:42 UTC
263c38d74d87b196ad08449002fc7799e76bad9d
1 file changed
+2
-1
.buildkite/test-pipeline.yaml
+2
-1
@@ -350,7 +350,8 @@ steps:
350
timeout_in_minutes: 25
351
gpu: h100
352
source_file_dependencies:
353
- - vllm/
353
+ - vllm/v1/attention
354
+ - vllm/model_executor/layers
355
- tests/v1/determinism/
356
commands:
357
- export VLLM_WORKER_MULTIPROC_METHOD=spawn