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 10, 21:30 UTC
[CI] fix test_chat_truncation_content_not_null test (#31488)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
Chauncey committed
Dec 29, 2025 at 20:47 UTC
48d5ca4e8b8b66dd0e734821d57dfc0eefaad4d2
1 file changed
+8
-2
tests/entrypoints/openai/test_response_api_with_harmony.py
+8
-2
@@ -999,9 +999,15 @@ async def test_function_call_with_previous_input_messages(
999
async def test_chat_truncation_content_not_null(client: OpenAI, model_name: str):
1000
response = await client.chat.completions.create(
1001
model=model_name,
1002
- messages=[{"role": "user", "content": "What is the role of AI in medicine?"}],
1002
+ messages=[
1003
+ {
1004
+ "role": "user",
1005
+ "content": "What is the role of AI in medicine?"
1006
+ "The response must exceed 350 words.",
1007
+ }
1008
+ ],
1009
temperature=0.0,
1004
- max_tokens=250,
1010
+ max_tokens=350,
1011
)
1012
1013
choice = response.choices[0]