| # # This tests the litelm proxy | |
| # # it makes async Completion requests with streaming | |
| # import openai | |
| # openai.base_url = "http://0.0.0.0:8000" | |
| # openai.api_key = "temp-key" | |
| # print(openai.base_url) | |
| # async def test_async_completion(): | |
| # response = await ( | |
| # model="gpt-3.5-turbo", | |
| # prompt='this is a test request, write a short poem', | |
| # ) | |
| # print(response) | |
| # print("test_streaming") | |
| # response = await openai.chat.completions.create( | |
| # model="gpt-3.5-turbo", | |
| # prompt='this is a test request, write a short poem', | |
| # stream=True | |
| # ) | |
| # print(response) | |
| # async for chunk in response: | |
| # print(chunk) | |
| # import asyncio | |
| # asyncio.run(test_async_completion()) | |