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, 03:15 UTC
[ROCm][CI] Split Entrypoints Integration (API Server 1) into 3 jobs (#37906)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Andreas Karatzas committed
Mar 23, 2026 at 20:48 UTC
de99d91ecee85b5d886ca0a608bfb69c2fcb8edb
1 file changed
+70
-4
.buildkite/test-amd.yaml
+70
-4
@@ -1435,7 +1435,7 @@ steps:
1435
- pytest -v -s entrypoints/offline_mode
1436
1437
1438
-- label: Entrypoints Integration (API Server 1) # 1h 7m
1438
+- label: Entrypoints Integration (API Server openai - Part 1) # TBD
1439
timeout_in_minutes: 180
1440
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi325]
1441
agent_pool: mi325_1
@@ -1448,10 +1448,43 @@ steps:
1448
- tests/entrypoints/test_chat_utils
1449
commands:
1450
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
1451
- - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/chat_completion/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/chat_completion/test_oot_registration.py --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py --ignore=entrypoints/openai/correctness/ --ignore=entrypoints/openai/tool_parsers/ --ignore=entrypoints/openai/responses
1451
+ - pytest -v -s entrypoints/openai/chat_completion --ignore=entrypoints/openai/chat_completion/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/chat_completion/test_oot_registration.py
1452
+
1453
+
1454
+- label: Entrypoints Integration (API Server openai - Part 2) # TBD
1455
+ timeout_in_minutes: 180
1456
+ mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi325]
1457
+ agent_pool: mi325_1
1458
+ fast_check: true
1459
+ torch_nightly: true
1460
+ working_dir: "/vllm-workspace/tests"
1461
+ source_file_dependencies:
1462
+ - vllm/
1463
+ - tests/entrypoints/openai
1464
+ - tests/entrypoints/test_chat_utils
1465
+ commands:
1466
+ - export VLLM_WORKER_MULTIPROC_METHOD=spawn
1467
+ - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
1468
+ - pytest -v -s entrypoints/openai/speech_to_text/
1469
- pytest -v -s entrypoints/test_chat_utils.py
1470
1471
1472
+- label: Entrypoints Integration (API Server openai - Part 3) # TBD
1473
+ timeout_in_minutes: 180
1474
+ mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi325]
1475
+ agent_pool: mi325_1
1476
+ fast_check: true
1477
+ torch_nightly: true
1478
+ working_dir: "/vllm-workspace/tests"
1479
+ source_file_dependencies:
1480
+ - vllm/
1481
+ - tests/entrypoints/openai
1482
+ - tests/entrypoints/test_chat_utils
1483
+ commands:
1484
+ - export VLLM_WORKER_MULTIPROC_METHOD=spawn
1485
+ - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/chat_completion --ignore=entrypoints/openai/completion --ignore=entrypoints/openai/speech_to_text/ --ignore=entrypoints/openai/correctness/ --ignore=entrypoints/openai/tool_parsers/ --ignore=entrypoints/openai/responses
1486
+
1487
+
1488
- label: Entrypoints Integration (API Server 2) #26.9m
1489
timeout_in_minutes: 45
1490
mirror_hardwares: [amdexperimental, amdproduction, amdgfx942nightly, amdmi325]
@@ -2941,7 +2974,7 @@ steps:
2974
# #
2975
#####################################################################################################################################
2976
2944
-- label: Entrypoints Integration (API Server 1) # TBD
2977
+- label: Entrypoints Integration (API Server openai - Part 1) # TBD
2978
timeout_in_minutes: 180
2979
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
2980
agent_pool: mi355_1
@@ -2954,10 +2987,43 @@ steps:
2987
- tests/entrypoints/test_chat_utils
2988
commands:
2989
- export VLLM_WORKER_MULTIPROC_METHOD=spawn
2957
- - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/chat_completion/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/chat_completion/test_oot_registration.py --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py --ignore=entrypoints/openai/correctness/ --ignore=entrypoints/openai/tool_parsers/ --ignore=entrypoints/openai/responses
2990
+ - pytest -v -s entrypoints/openai/chat_completion --ignore=entrypoints/openai/chat_completion/test_chat_with_tool_reasoning.py --ignore=entrypoints/openai/chat_completion/test_oot_registration.py
2991
+
2992
+
2993
+- label: Entrypoints Integration (API Server openai - Part 2) # TBD
2994
+ timeout_in_minutes: 180
2995
+ mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
2996
+ agent_pool: mi355_1
2997
+ fast_check: true
2998
+ torch_nightly: true
2999
+ working_dir: "/vllm-workspace/tests"
3000
+ source_file_dependencies:
3001
+ - vllm/
3002
+ - tests/entrypoints/openai
3003
+ - tests/entrypoints/test_chat_utils
3004
+ commands:
3005
+ - export VLLM_WORKER_MULTIPROC_METHOD=spawn
3006
+ - pytest -v -s entrypoints/openai/completion --ignore=entrypoints/openai/completion/test_tensorizer_entrypoint.py
3007
+ - pytest -v -s entrypoints/openai/speech_to_text/
3008
- pytest -v -s entrypoints/test_chat_utils.py
3009
3010
3011
+- label: Entrypoints Integration (API Server openai - Part 3) # TBD
3012
+ timeout_in_minutes: 180
3013
+ mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]
3014
+ agent_pool: mi355_1
3015
+ fast_check: true
3016
+ torch_nightly: true
3017
+ working_dir: "/vllm-workspace/tests"
3018
+ source_file_dependencies:
3019
+ - vllm/
3020
+ - tests/entrypoints/openai
3021
+ - tests/entrypoints/test_chat_utils
3022
+ commands:
3023
+ - export VLLM_WORKER_MULTIPROC_METHOD=spawn
3024
+ - pytest -v -s entrypoints/openai --ignore=entrypoints/openai/chat_completion --ignore=entrypoints/openai/completion --ignore=entrypoints/openai/speech_to_text/ --ignore=entrypoints/openai/correctness/ --ignore=entrypoints/openai/tool_parsers/ --ignore=entrypoints/openai/responses
3025
+
3026
+
3027
- label: Entrypoints Integration (API Server 2) # TBD
3028
timeout_in_minutes: 180
3029
mirror_hardwares: [amdexperimental, amdproduction, amdgfx950nightly, amdmi355]