[V1] V1 Enablement Oracle (#13726)

Signed-off-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: Michael Goin <michael@neuralmagic.com>
This commit is contained in:
Robert Shaw
2025-03-15 01:02:20 -04:00
committed by GitHub
parent 8c0d15d5c5
commit d4d93db2c5
96 changed files with 1537 additions and 512 deletions

View File

@@ -3,12 +3,21 @@
import sys
from contextlib import nullcontext
import pytest
from vllm_test_utils import BlameResult, blame
from vllm import LLM, SamplingParams
from vllm.distributed import cleanup_dist_env_and_memory
@pytest.fixture(scope="function", autouse=True)
def use_v0_only(monkeypatch):
"""
V1 only supports xgrammar so this is irrelevant.
"""
monkeypatch.setenv('VLLM_USE_V1', '0')
def run_normal_opt125m():
prompts = [
"Hello, my name is",

View File

@@ -10,7 +10,6 @@ from ...utils import RemoteOpenAIServer
# # any model with a chat template should work here
MODEL_NAME = "Qwen/Qwen2-1.5B-Instruct"
DUMMY_CHAT_TEMPLATE = """{% for message in messages %}{{message['role'] + ': ' + message['content'] + '\\n'}}{% endfor %}""" # noqa: E501
@pytest.fixture(scope="module")
@@ -22,8 +21,6 @@ def server():
"--enforce-eager",
"--max-model-len",
"4080",
"--chat-template",
DUMMY_CHAT_TEMPLATE,
]
with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:

View File

@@ -11,7 +11,6 @@ from ...utils import RemoteOpenAIServer
# # any model with a chat template should work here
MODEL_NAME = "Qwen/Qwen2-1.5B-Instruct"
DUMMY_CHAT_TEMPLATE = """{% for message in messages %}{{message['role'] + ': ' + message['content'] + '\\n'}}{% endfor %}""" # noqa: E501
API_KEY = "abc-123"
ERROR_API_KEY = "abc"
ROOT_PATH = "llm"
@@ -28,8 +27,6 @@ def server():
"4080",
"--root-path", # use --root-path=/llm for testing
"/" + ROOT_PATH,
"--chat-template",
DUMMY_CHAT_TEMPLATE,
]
envs = os.environ.copy()