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

Use smaller embedding model when not testing model specifically (#13891)

Browse Source
This commit is contained in:
Harry Mellor
2025-02-28 08:50:43 +00:00
committed by GitHub
parent b9e41734c5
commit 76c89fcadd
9 changed files with 15 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/entrypoints/openai/test_metrics.py
View File

@@ -282,7 +282,7 @@ 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/e5-mistral-7b-instruct", "input": "You are a helpful assistant."}}""" # noqa: E501
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"
port = "8001"
@@ -302,7 +302,7 @@ def test_metrics_exist_run_batch(use_v1: bool):
"-o",
output_file.name,
"--model",
"intfloat/e5-mistral-7b-instruct",
"intfloat/multilingual-e5-small",
"--enable-metrics",
"--url",
base_url,