litellm-mirror/tests/llm_translation/test_aiohttp_openai.py
Ishaan Jaff a85de46ef7
(proxy - RPS) - Get 2K RPS at 4 instances, minor fix aiohttp_openai/ (#7659)
* speed up transform_response

* use 2 workers

* undo changes to uvicorn

* ci/cd run again
2025-01-09 17:24:18 -08:00

33 lines
851 B
Python

import json
import os
import sys
from datetime import datetime
import pytest
sys.path.insert(
0, os.path.abspath("../../")
) # Adds the parent directory to the system-path
import litellm
@pytest.mark.asyncio()
async def test_aiohttp_openai():
litellm.set_verbose = True
response = await litellm.acompletion(
model="aiohttp_openai/fake-model",
messages=[{"role": "user", "content": "Hello, world!"}],
api_base="https://exampleopenaiendpoint-production.up.railway.app/v1/chat/completions",
api_key="fake-key",
)
print(response)
@pytest.mark.asyncio()
async def test_aiohttp_openai_gpt_4o():
litellm.set_verbose = True
response = await litellm.acompletion(
model="aiohttp_openai/gpt-4o",
messages=[{"role": "user", "content": "Hello, world!"}],
)
print(response)