@tduncan / Nvidia-Optimization-Stack-VLLM / commits / 23cfaad497

[CI] Update entrypoints CI (#56763)

Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>

wang.yuqi committed Sep 14, 2026 at 15:59 UTC 23cfaad49701c497def53552b23317335431f72a
2 files changed +13 -13
.buildkite/test-amd.yaml
+6 -6
@@ -1067,7 +1067,7 @@ steps:
1067 - PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/serve/dev/rpc
1068 - pytest -v -s entrypoints/scale_out
1069
1070 -- label: ":amd: (MI300) Entrypoints Integration (API Server OpenAI - Part 1)" # TBD
1070 +- label: ":amd: (MI300) Entrypoints Integration (OpenAI API completion)" # TBD
1071 timeout_in_minutes: 40
1072 mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
1073 dind: false
@@ -1080,9 +1080,10 @@ steps:
1080 - tests/entrypoints/openai
1081 commands:
1082 - export VLLM_WORKER_MULTIPROC_METHOD=spawn
1083 + - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
1084 - pytest -v -s entrypoints/openai/ --ignore=entrypoints/openai/completion --ignore=entrypoints/openai/chat_completion --ignore=entrypoints/openai/responses --ignore=entrypoints/openai/correctness
1085
1085 -- label: ":amd: (MI300) Entrypoints Integration (API Server OpenAI - Part 2)" # TBD
1086 +- label: ":amd: (MI300) Entrypoints Integration (OpenAI API chat_completion)" # TBD
1087 timeout_in_minutes: 55
1088 mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi300]
1089 dind: false
@@ -1096,7 +1097,6 @@ steps:
1097 commands:
1098 - export VLLM_WORKER_MULTIPROC_METHOD=spawn
1099 - pytest -v -s entrypoints/openai/chat_completion
1099 - - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
1100
1101 - label: ":amd: (MI300) Entrypoints Integration (API Server Generate)" # TBD
1102 timeout_in_minutes: 65
@@ -3217,7 +3217,7 @@ steps:
3217 - PYTHONPATH=/vllm-workspace pytest -v -s entrypoints/serve/dev/rpc
3218 - pytest -v -s entrypoints/scale_out
3219
3220 -- label: ":amd: (MI355) Entrypoints Integration (API Server OpenAI - Part 1)" # TBD
3220 +- label: ":amd: (MI355) Entrypoints Integration (OpenAI API completion)" # TBD
3221 timeout_in_minutes: 35
3222 mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
3223 dind: false
@@ -3230,9 +3230,10 @@ steps:
3230 - tests/entrypoints/openai
3231 commands:
3232 - export VLLM_WORKER_MULTIPROC_METHOD=spawn
3233 + - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
3234 - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/completion --ignore=entrypoints/openai/chat_completion --ignore=entrypoints/openai/responses --ignore=entrypoints/openai/correctness
3235
3235 -- label: ":amd: (MI355) Entrypoints Integration (API Server OpenAI - Part 2)" # TBD
3236 +- label: ":amd: (MI355) Entrypoints Integration (OpenAI API chat_completion)" # TBD
3237 timeout_in_minutes: 50
3238 mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
3239 dind: false
@@ -3246,7 +3247,6 @@ steps:
3247 commands:
3248 - export VLLM_WORKER_MULTIPROC_METHOD=spawn
3249 - pytest -v -s entrypoints/openai/chat_completion
3249 - - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
3250
3251 - label: ":amd: (MI355) Entrypoints Integration (API Server Generate)" # TBD
3252 timeout_in_minutes: 40
.buildkite/test_areas/entrypoints.yaml
+7 -7
@@ -73,9 +73,9 @@ steps:
73 depends_on:
74 - image-build-amd
75
76 -- label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (API Server OpenAI - Part 1)"
76 +- label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (OpenAI API completion)"
77 device: h200_35gb
78 - key: entrypoints-integration-api-server-openai-part-1
78 + key: entrypoints-integration-api-server-openai-completion
79 timeout_in_minutes: 68
80 working_dir: "/vllm-workspace/tests"
81 source_file_dependencies:
@@ -84,19 +84,20 @@ steps:
84 - tests/entrypoints/openai
85 commands:
86 - export VLLM_WORKER_MULTIPROC_METHOD=spawn
87 + - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
88 - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/completion --ignore=entrypoints/openai/chat_completion --ignore=entrypoints/openai/responses --ignore=entrypoints/openai/correctness
89 mirror:
90 amd:
90 - label: ":amd: (MI300) Entrypoints Integration (API Server OpenAI - Part 1)"
91 + label: ":amd: (MI300) Entrypoints Integration (OpenAI API completion)"
92 dind: false
93 device: mi300_1
94 timeout_in_minutes: 65
95 depends_on:
96 - image-build-amd
97
97 -- label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (API Server OpenAI - Part 2)"
98 +- label: ":nvidia: (H200 MIG 35GB) Entrypoints Integration (OpenAI API chat_completion)"
99 device: h200_35gb
99 - key: entrypoints-integration-api-server-openai-part-2
100 + key: entrypoints-integration-api-server-openai-chat_completion
101 timeout_in_minutes: 83
102 working_dir: "/vllm-workspace/tests"
103 source_file_dependencies:
@@ -106,10 +107,9 @@ steps:
107 commands:
108 - export VLLM_WORKER_MULTIPROC_METHOD=spawn
109 - pytest -v -s entrypoints/openai/chat_completion
109 - - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
110 mirror:
111 amd:
112 - label: ":amd: (MI300) Entrypoints Integration (API Server OpenAI - Part 2)"
112 + label: ":amd: (MI300) Entrypoints Integration (OpenAI API chat_completion)"
113 dind: false
114 device: mi300_1
115 timeout_in_minutes: 70