biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[CI][MCP][Harmony] Heavy refactoring Harmony & MCP response tests and stabilizing with deterministic test infrastructure (#33949)

Browse Source 
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
This commit is contained in:
Andreas Karatzas
2026-02-20 22:03:32 -06:00
committed by GitHub
parent 5719a4e4e6
commit 991d6bff38
10 changed files with 1187 additions and 886 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
tests/entrypoints/openai/responses/test_simple.py
View File

@@ -12,13 +12,15 @@ MODEL_NAME = "Qwen/Qwen3-8B"
@pytest.fixture(scope="module")
def server():
args = ["--reasoning-parser", "qwen3", "--max_model_len", "5000"]
env_dict = dict(
VLLM_ENABLE_RESPONSES_API_STORE="1",
# uncomment for tool calling
# PYTHON_EXECUTION_BACKEND="dangerously_use_uv",
)
from .conftest import BASE_TEST_ENV
args = ["--reasoning-parser", "qwen3", "--max_model_len", "5000"]
env_dict = {
**BASE_TEST_ENV,
"VLLM_ENABLE_RESPONSES_API_STORE": "1",
# uncomment for tool calling
# PYTHON_EXECUTION_BACKEND: "dangerously_use_uv",
}
with RemoteOpenAIServer(MODEL_NAME, args, env_dict=env_dict) as remote_server:
yield remote_server