async-openai
async-openai copied to clipboard
Unofficial Async Python client library for the OpenAI API
This library appears to have pretty exhaustive type coverage, is there any change you can mark it with a [py.typed](https://pyo3.rs/v0.20.2/python_typing_hints#pytyped-requirement) which will make it [PEP561](https://www.python.org/dev/peps/pep-0561/) complaint and stop tools like...
After installing 'async-openai' into my projects environment with pipenv, I am receiving this issue: ``` In [1]: from async_openai import OpenAI, settings, CompletionResponse .... 294 return import_string(REDIRECT_TO_V1[import_path]) 295 if import_path...
error `"cannot pickle 'classmethod' object"` occurs when importing [log (2).txt](https://github.com/GrowthEngineAI/async-openai/files/12922039/log.2.txt)
Please include some examples in your document, if this package supports concurrent calls. Thanks!
You need better documentation. I am surprised you cant add model or even endpoint, and there is no any info about how to do it,
async_openai.types.errors.InvalidRequestError: -1 is less than the minimum of 0 - 'max_tokens'
Hello. Currently, there is no messages parameter for async completions. Is there any other way to do this?
```py File "/root/TelegramAIBot/main.py", line 2, in from async_openai import OpenAI, CompletionResponse File "/root/TelegramAIBot/venv/lib/python3.10/site-packages/async_openai/__init__.py", line 26, in from async_openai.schemas.completions import CompletionChoice, CompletionObject, CompletionResponse File "/root/TelegramAIBot/venv/lib/python3.10/site-packages/async_openai/schemas/__init__.py", line 3, in from async_openai.schemas.completions import...