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 6, 08:15 UTC
[Refactor] Move serve entrypoint tests under tests/entrypoints/serve/ (#37595)
Signed-off-by: sfeng33 <4florafeng@gmail.com>
Flora Feng committed
Mar 20, 2026 at 05:10 UTC
6050b93bedb66a086281c160814b3cfca8da3111
11 files changed
+13
-14
.buildkite/test-amd.yaml
+7
-7
@@ -223,7 +223,7 @@ steps:
223
- vllm/platforms/rocm.py
224
commands:
225
- pytest -v -s entrypoints/openai/tool_parsers
226
- - pytest -v -s entrypoints/ --ignore=entrypoints/llm --ignore=entrypoints/rpc --ignore=entrypoints/sleep --ignore=entrypoints/instrumentator --ignore=entrypoints/openai --ignore=entrypoints/offline_mode --ignore=entrypoints/test_chat_utils.py --ignore=entrypoints/pooling
226
+ - pytest -v -s entrypoints/ --ignore=entrypoints/llm --ignore=entrypoints/rpc --ignore=entrypoints/sleep --ignore=entrypoints/serve/instrumentator --ignore=entrypoints/openai --ignore=entrypoints/offline_mode --ignore=entrypoints/test_chat_utils.py --ignore=entrypoints/pooling
227
228
229
- label: Entrypoints Integration (LLM) # TBD
@@ -254,11 +254,11 @@ steps:
254
source_file_dependencies:
255
- vllm/
256
- tests/entrypoints/rpc
257
- - tests/entrypoints/instrumentator
257
+ - tests/entrypoints/serve/instrumentator
258
- tests/tool_use
259
commands:
260
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
261
- - pytest -v -s entrypoints/instrumentator
261
+ - pytest -v -s entrypoints/serve/instrumentator
262
- PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/rpc
263
- pytest -v -s tool_use
264
@@ -1475,11 +1475,11 @@ steps:
1475
source_file_dependencies:
1476
- vllm/
1477
- tests/entrypoints/rpc
1478
- - tests/entrypoints/instrumentator
1478
+ - tests/entrypoints/serve/instrumentator
1479
- tests/tool_use
1480
commands:
1481
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
1482
- - pytest -v -s entrypoints/instrumentator
1482
+ - pytest -v -s entrypoints/serve/instrumentator
1483
- PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/rpc
1484
- pytest -v -s tool_use
1485
@@ -2981,11 +2981,11 @@ steps:
2981
source_file_dependencies:
2982
- vllm/
2983
- tests/entrypoints/rpc
2984
- - tests/entrypoints/instrumentator
2984
+ - tests/entrypoints/serve/instrumentator
2985
- tests/tool_use
2986
commands:
2987
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
2988
- - pytest -v -s entrypoints/instrumentator
2988
+ - pytest -v -s entrypoints/serve/instrumentator
2989
- PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/rpc
2990
- pytest -v -s tool_use
2991
.buildkite/test_areas/entrypoints.yaml
+3
-3
@@ -10,7 +10,7 @@ steps:
10
- tests/entrypoints/
11
commands:
12
- pytest -v -s entrypoints/openai/tool_parsers
13
- - pytest -v -s entrypoints/ --ignore=entrypoints/llm --ignore=entrypoints/rpc --ignore=entrypoints/sleep --ignore=entrypoints/instrumentator --ignore=entrypoints/openai --ignore=entrypoints/offline_mode --ignore=entrypoints/test_chat_utils.py --ignore=entrypoints/pooling
13
+ - pytest -v -s entrypoints/ --ignore=entrypoints/llm --ignore=entrypoints/rpc --ignore=entrypoints/sleep --ignore=entrypoints/serve/instrumentator --ignore=entrypoints/openai --ignore=entrypoints/offline_mode --ignore=entrypoints/test_chat_utils.py --ignore=entrypoints/pooling
14
15
- label: Entrypoints Integration (LLM)
16
timeout_in_minutes: 40
@@ -48,11 +48,11 @@ steps:
48
source_file_dependencies:
49
- vllm/
50
- tests/entrypoints/rpc
51
- - tests/entrypoints/instrumentator
51
+ - tests/entrypoints/serve/instrumentator
52
- tests/tool_use
53
commands:
54
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
55
- - pytest -v -s entrypoints/instrumentator
55
+ - pytest -v -s entrypoints/serve/instrumentator
56
- PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/rpc
57
- pytest -v -s tool_use
58
tests/entrypoints/openai/cpu/__init__.py
tests/entrypoints/serve/instrumentator/__init__.py
renamed
tests/entrypoints/serve/instrumentator/test_basic.py
renamed
+1
-2
@@ -11,11 +11,10 @@ import pytest_asyncio
11
import requests
12
from fastapi import Request
13
14
+from tests.utils import RemoteOpenAIServer
15
from vllm.v1.engine.exceptions import EngineDeadError
16
from vllm.version import __version__ as VLLM_VERSION
17
17
-from ...utils import RemoteOpenAIServer
18
-
18
MODEL_NAME = "Qwen/Qwen3-0.6B"
19
20
tests/entrypoints/serve/instrumentator/test_metrics.py
renamed
tests/entrypoints/serve/instrumentator/test_optional_middleware.py
renamed
+1
-1
@@ -10,7 +10,7 @@ from http import HTTPStatus
10
import pytest
11
import requests
12
13
-from ...utils import RemoteOpenAIServer
13
+from tests.utils import RemoteOpenAIServer
14
15
# Use a small embeddings model for faster startup and smaller memory footprint.
16
# Since we are not testing any chat functionality,
tests/entrypoints/serve/instrumentator/test_orca_metrics.py
renamed
+1
-1
@@ -5,7 +5,7 @@ import openai
5
import pytest
6
import pytest_asyncio
7
8
-from ...utils import RemoteOpenAIServer
8
+from tests.utils import RemoteOpenAIServer
9
10
# any model with a chat template should work here
11
MODEL_NAME = "Qwen/Qwen3-0.6B"