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 15, 19:45 UTC
[ci] Reduce number of tests in fastcheck (#14782)
Kevin H. Luu committed
Mar 13, 2025 at 20:43 UTC
f1f632d9eca4967f25d2f09a8b444724d2cda624
1 file changed
-5
.buildkite/test-pipeline.yaml
-5
@@ -41,7 +41,6 @@ steps:
41
- grep \"sig sig-object py\" build/html/api/inference_params.html
42
43
- label: Async Engine, Inputs, Utils, Worker Test # 24min
44
- fast_check: true
44
source_file_dependencies:
45
- vllm/
46
- tests/mq_llm_engine
@@ -126,7 +125,6 @@ steps:
125
- label: Distributed Tests (4 GPUs) # 10min
126
working_dir: "/vllm-workspace/tests"
127
num_gpus: 4
129
- fast_check: true
128
source_file_dependencies:
129
- vllm/distributed/
130
- vllm/core/
@@ -152,7 +150,6 @@ steps:
150
151
- label: Metrics, Tracing Test # 10min
152
num_gpus: 2
155
- fast_check: true
153
source_file_dependencies:
154
- vllm/
155
- tests/metrics
@@ -284,7 +281,6 @@ steps:
281
parallelism: 4
282
283
- label: PyTorch Fullgraph Smoke Test # 9min
287
- fast_check: true
284
source_file_dependencies:
285
- vllm/
286
- tests/compile
@@ -528,7 +524,6 @@ steps:
524
- label: Plugin Tests (2 GPUs) # 40min
525
working_dir: "/vllm-workspace/tests"
526
num_gpus: 2
531
- fast_check: true
527
source_file_dependencies:
528
- vllm/plugins/
529
- tests/plugins/