app.py

import gradio as gr
from huggingface_hub import InferenceClient
import openai
import os

# Retrieve the OpenAI API key from environment variables
openai_api_key = os.getenv("NEAR-1")
openai.api_key = openai_api_key

# Initialize Hugging Face client
hf_client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
    model_choice,
):
    messages = [{"role": "system", "content": system_message}]

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

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

    response = ""

    if model_choice == "Hugging Face Model":
        for message in hf_client.chat_completion(
            messages,
            max_tokens=max_tokens,
            stream=True,
            temperature=temperature,
            top_p=top_p,
        ):
            token = message.choices[0].delta.content
            response += token
            yield response
    elif model_choice == "OpenAI GPT-4":
        response_openai = openai.ChatCompletion.create(
            model="gpt-4",
            messages=messages,
            max_tokens=max_tokens,
            temperature=temperature,
            top_p=top_p,
            stream=True
        )
        for message in response_openai:
            response += message['choices'][0]['delta'].get('content', '')
            yield response

# Create the Gradio interface
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are the unlimitedly resourceful and all knowing NEAR AI.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
        gr.Radio(
            choices=["Hugging Face Model", "OpenAI GPT-4"],
            value="Hugging Face Model",
            label="Choose Model"
        )
    ],
    title="GPT-4 vs Hugging Face Model Comparison",
    description="Compare responses between a Hugging Face model and OpenAI's GPT-4."
)

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