gradio_openai.py

import os
import logging

import gradio as gr
from openai import OpenAI
from dotenv import load_dotenv

load_dotenv(".env")

logging.basicConfig(level=logging.INFO)
logging.getLogger("gradio").setLevel(logging.INFO)
logging.getLogger("httpx").setLevel(logging.WARNING)


client = OpenAI()


def generate_completion(input, history):

    messages = [
        {
            "role": "system",
            "content": "You are a world-class assistant.",
        }
    ]

    if history:
        for entry in history:
            if len(entry) == 2:
                messages.append(
                    {
                        "role": "user",
                        "content": entry[0],
                    }
                )
                messages.append(
                    {
                        "role": "assistant",
                        "content": entry[1],
                    }
                )

    messages.append(
        {
            "role": "user",
            "content": input,
        }
    )

    response = client.chat.completions.create(
        model="gpt-3.5-turbo-0125",
        messages=messages,  # type: ignore
        stream=True,
        temperature=0,
        max_tokens=4000,
    )  # type: ignore

    answer_str: str = ""
    for chunk in response:
        if chunk.choices[0].delta.content is not None:
            answer_str += chunk.choices[0].delta.content
        else:
            answer_str += ""
        yield answer_str


demo = gr.ChatInterface(fn=generate_completion)

if __name__ == "__main__":
    demo.queue()
    demo.launch()