mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2025-12-24 06:15:01 +08:00
37 lines
1.2 KiB
Python
37 lines
1.2 KiB
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
|
|
import pytest
|
|
|
|
from vllm import LLM, SamplingParams
|
|
|
|
from ...utils import create_new_process_for_each_test
|
|
|
|
|
|
@create_new_process_for_each_test()
|
|
@pytest.mark.parametrize("attn_backend", ["FLASH_ATTN", "FLASHINFER"])
|
|
def test_cascade_attention(example_system_message, attn_backend):
|
|
prompt = "\n<User>: Implement fibonacci sequence in Python.\n<Claude>:"
|
|
|
|
if attn_backend == "FLASHINFER":
|
|
pytest.skip(
|
|
"This test is failing with FlashInfer backend and "
|
|
"needs investigation. See issue #25679."
|
|
)
|
|
|
|
llm = LLM(
|
|
model="Qwen/Qwen2-1.5B-Instruct", attention_config={"backend": attn_backend}
|
|
)
|
|
sampling_params = SamplingParams(temperature=0.0, max_tokens=100)
|
|
|
|
# No cascade attention.
|
|
single_prompt = [example_system_message + prompt]
|
|
responses = llm.generate(single_prompt, sampling_params)
|
|
ref_output = responses[0].outputs[0].text
|
|
|
|
# (Probably) Use cascade attention.
|
|
prompts = [example_system_message + prompt] * 64
|
|
responses = llm.generate(prompts, sampling_params)
|
|
for response in responses:
|
|
assert response.outputs[0].text == ref_output
|