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 12, 06:00 UTC
[ci] Make distributed 8 gpus test optional (#29801)
Signed-off-by: Kevin H. Luu <khluu000@gmail.com>
Kevin H. Luu committed
Dec 1, 2025 at 10:22 UTC
ec7035c9d40cda43fa9c3d2d1c148b1382aa04a4
1 file changed
+2
-1
.buildkite/test-pipeline.yaml
+2
-1
@@ -215,6 +215,7 @@ steps:
215
timeout_in_minutes: 10
216
gpu: h100
217
num_gpus: 8
218
+ optional: true
219
working_dir: "/vllm-workspace/tests"
220
source_file_dependencies:
221
- examples/offline_inference/torchrun_dp_example.py
@@ -1369,4 +1370,4 @@ steps:
1370
num_gpus: 2
1371
working_dir: "/vllm-workspace"
1372
commands:
1372
- - bash .buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_block_ep_eplb.sh 0.8 200 8020 2 1
\ No newline at end of file
1373
+ - bash .buildkite/scripts/scheduled_integration_test/qwen30b_a3b_fp8_block_ep_eplb.sh 0.8 200 8020 2 1