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, 18:30 UTC
[V1] Enable Entrypoints Tests (#14903)
Robert Shaw committed
Mar 16, 2025 at 20:56 UTC
aecc780dba30db6b503754926564642374cb2c2e
2 files changed
+4
.buildkite/test-pipeline.yaml
+1
@@ -198,6 +198,7 @@ steps:
198
commands:
199
# split the test to avoid interference
200
- pytest -v -s v1/core
201
+ - pytest -v -s v1/entrypoints
202
- pytest -v -s v1/engine
203
- pytest -v -s v1/sample
204
- pytest -v -s v1/worker
tests/v1/entrypoints/llm/test_struct_output_generate.py
+3
@@ -18,6 +18,9 @@ MODELS_TO_TEST = [
18
"Qwen/Qwen2.5-1.5B-Instruct", "mistralai/Ministral-8B-Instruct-2410"
19
]
20
21
+# Undo after https://github.com/vllm-project/vllm/pull/14868
22
+pytest.skip(allow_module_level=True)
23
+
24
25
@pytest.mark.skip_global_cleanup
26
@pytest.mark.parametrize("guided_decoding_backend",