49 lines
1.4 KiB
Python
49 lines
1.4 KiB
Python
import pytest
|
|
from openai import OpenAI
|
|
from utils import *
|
|
|
|
server: ServerProcess
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def create_server():
|
|
global server
|
|
server = ServerPreset.tinyllama2()
|
|
|
|
def test_responses_with_openai_library():
|
|
global server
|
|
server.start()
|
|
client = OpenAI(api_key="dummy", base_url=f"http://{server.server_host}:{server.server_port}/v1")
|
|
res = client.responses.create(
|
|
model="gpt-4.1",
|
|
input=[
|
|
{"role": "system", "content": "Book"},
|
|
{"role": "user", "content": "What is the best book"},
|
|
],
|
|
max_output_tokens=8,
|
|
temperature=0.8,
|
|
)
|
|
assert match_regex("(Suddenly)+", res.output_text)
|
|
|
|
def test_responses_stream_with_openai_library():
|
|
global server
|
|
server.start()
|
|
client = OpenAI(api_key="dummy", base_url=f"http://{server.server_host}:{server.server_port}/v1")
|
|
stream = client.responses.create(
|
|
model="gpt-4.1",
|
|
input=[
|
|
{"role": "system", "content": "Book"},
|
|
{"role": "user", "content": "What is the best book"},
|
|
],
|
|
max_output_tokens=8,
|
|
temperature=0.8,
|
|
stream=True,
|
|
)
|
|
|
|
gathered_text = ''
|
|
for r in stream:
|
|
if r.type == "response.output_text.delta":
|
|
gathered_text += r.delta
|
|
if r.type == "response.completed":
|
|
assert gathered_text == r.response.output_text
|
|
assert match_regex("(Suddenly)+", r.response.output_text)
|