biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Frontend] run-batch supports V1 (#21541)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-07-25 11:05:55 +08:00
committed by GitHub
parent fe56180c7f
commit 34ddcf9ff4
5 changed files with 56 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
tests/entrypoints/openai/test_metrics.py
View File

@@ -295,8 +295,6 @@ async def test_metrics_exist(server: RemoteOpenAIServer,
def test_metrics_exist_run_batch(use_v1: bool):
if use_v1:
pytest.skip("Skipping test on vllm V1")
input_batch = """{"custom_id": "request-0", "method": "POST", "url": "/v1/embeddings", "body": {"model": "intfloat/multilingual-e5-small", "input": "You are a helpful assistant."}}""" # noqa: E501
base_url = "0.0.0.0"
@@ -323,7 +321,8 @@ def test_metrics_exist_run_batch(use_v1: bool):
base_url,
"--port",
port,
], )
],
env={"VLLM_USE_V1": "1" if use_v1 else "0"})
def is_server_up(url):
try: