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 13, 20:45 UTC
[CI/Build] Remove the flaky gpt-oss lora test (#27966)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Jee Jee Li committed
Nov 3, 2025 at 16:50 UTC
32257297dd4dcb996a0fb4641c2018289d20396b
1 file changed
-5
tests/lora/test_gptoss_tp.py
-5
@@ -32,7 +32,6 @@ The Competition_ID of competition_record is the foreign key of Competition_ID of
32
###Response:<|end|><|start|>assistant<|channel|>final<|message|>""" # noqa: E501
33
34
EXPECTED_LORA_OUTPUT = [
35
- "SELECT AVG(Working_Horses) FROM farm WHERE Total_Horses > 5000;",
35
"SELECT AVG(Working_Horses) FROM farm WHERE Total_Horses > 5000;",
36
"SELECT MAX(Cows) AS Max_Cows, MIN(Cows) AS Min_Cows FROM farm;",
37
"SELECT MAX(Cows) AS Max_Cows, MIN(Cows) AS Min_Cows FROM farm;",
@@ -41,9 +40,6 @@ EXPECTED_LORA_OUTPUT = [
40
41
def generate_and_test(llm: vllm.LLM, lora_path: str, lora_id: int) -> None:
42
prompts = [
44
- PROMPT_TEMPLATE.format(
45
- context="What is the average number of working horses of farms with more than 5000 total number of horses?" # noqa: E501
46
- ), # noqa: E501
43
PROMPT_TEMPLATE.format(
44
context="Give the average number of working horses on farms with more than 5000 total horses." # noqa: E501
45
), # noqa: E501
@@ -67,7 +63,6 @@ def generate_and_test(llm: vllm.LLM, lora_path: str, lora_id: int) -> None:
63
generated_text = output.outputs[0].text.strip()
64
generated_texts.append(generated_text)
65
print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
70
-
66
for i in range(len(EXPECTED_LORA_OUTPUT)):
67
assert generated_texts[i].startswith(EXPECTED_LORA_OUTPUT[i])
68