aphrodite-engine/tests/prefix_caching/test_disable_sliding_window.py

"""Compare the with and without prefix caching.

Run `pytest tests/prefix_caching/test_prefix_caching.py`.
"""
import pytest

from aphrodite import LLM
from tests.conftest import cleanup

MODEL_LEN_LEN = [
    # Example models with sliding window.
    ("bigcode/starcoder2-3b", 4096, 16384),
    # ("mistralai/Mistral-7B-v0.1", 4096, 32768), << OOM in CI

    # Confirm model with sliding window works.
    # config has "use_sliding_window": false
    ("Qwen/Qwen1.5-0.5B-Chat", 32768, 32768),
    # config has no sliding window attribute.
    ("TinyLlama/TinyLlama-1.1B-Chat-v1.0", 2048, 2048),
]


@pytest.mark.parametrize("model_len_len", MODEL_LEN_LEN)
def test_disable_sliding_window(model_len_len, ):
    model, sliding_len, full_len = model_len_len
    aphrodite_disabled_model = LLM(model, disable_sliding_window=True)
    aphrodite_disabled_model.generate("Hi my name is")
    model_config = aphrodite_disabled_model.llm_engine.model_config
    assert model_config.max_model_len == sliding_len, (
        "Max len expected to equal sliding_len of %s, but got %s", sliding_len,
        model_config.max_model_len)

    del aphrodite_disabled_model
    cleanup()

    aphrodite_enabled_model = LLM(model, disable_sliding_window=False)
    aphrodite_enabled_model.generate("Hi my name is")
    model_config = aphrodite_enabled_model.llm_engine.model_config
    assert model_config.max_model_len == full_len, (
        "Max len expected to equal full_len of %s, but got %s", full_len,
        model_config.max_model_len)

    del aphrodite_enabled_model
    cleanup()