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, 02:15 UTC
[CI/Build] Split Entrypoints Test into LLM and API Server (#20945)
Signed-off-by: mgoin <mgoin64@gmail.com>
Michael Goin committed
Jul 15, 2025 at 11:44 UTC
946aadb4a0e07bf735b0f0145230c59002dc5089
1 file changed
+14
-4
.buildkite/test-pipeline.yaml
+14
-4
@@ -117,7 +117,7 @@ steps:
117
commands:
118
- pytest -v -s core
119
120
-- label: Entrypoints Test # 40min
120
+- label: Entrypoints Test (LLM) # 40min
121
mirror_hardwares: [amdexperimental]
122
working_dir: "/vllm-workspace/tests"
123
fast_check: true
@@ -125,8 +125,6 @@ steps:
125
source_file_dependencies:
126
- vllm/
127
- tests/entrypoints/llm
128
- - tests/entrypoints/openai
129
- - tests/entrypoints/test_chat_utils
128
- tests/entrypoints/offline_mode
129
commands:
130
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
@@ -135,9 +133,21 @@ steps:
133
- pytest -v -s entrypoints/llm/test_generate.py # it needs a clean process
134
- pytest -v -s entrypoints/llm/test_generate_multiple_loras.py # it needs a clean process
135
- VLLM_USE_V1=0 pytest -v -s entrypoints/llm/test_guided_generate.py # it needs a clean process
136
+ - VLLM_USE_V1=0 pytest -v -s entrypoints/offline_mode # Needs to avoid interference with other tests
137
+
138
+- label: Entrypoints Test (API Server) # 40min
139
+ mirror_hardwares: [amdexperimental]
140
+ working_dir: "/vllm-workspace/tests"
141
+ fast_check: true
142
+ torch_nightly: true
143
+ source_file_dependencies:
144
+ - vllm/
145
+ - tests/entrypoints/openai
146
+ - tests/entrypoints/test_chat_utils
147
+ commands:
148
+ - export VLLM_WORKER_MULTIPROC_METHOD=spawn
149
- pytest -v -s entrypoints/openai --ignore=entrypoints/openai/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/test_oot_registration.py --ignore=entrypoints/openai/test_tensorizer_entrypoint.py --ignore=entrypoints/openai/correctness/
150
- pytest -v -s entrypoints/test_chat_utils.py
140
- - VLLM_USE_V1=0 pytest -v -s entrypoints/offline_mode # Needs to avoid interference with other tests
151
152
- label: Distributed Tests (4 GPUs) # 10min
153
mirror_hardwares: [amdexperimental]