app.py

import gradio as gr
import requests

def generate_message(prompt, model):
    url = 'https://api.together.xyz/inference'
    headers = {
        'Authorization': 'Bearer ee16b2ea8440ef7142ac5e21f5d096ee2dfb9c0c206acff4818a35b91ae59cef',
        'accept': 'application/json',
        'content-type': 'application/json'
    }
    data = {
        "model": model,
        "prompt": prompt,
        "max_tokens": 512,
        "stop": [
            "</s>",
            "[INST]",
            "<<SYS>>",
            "<</SYS>>",
            "[/INST]",
            "<s>"
        ],
        "temperature": 0.75,
        "top_p": 0.2,
        "top_k": 10
    }
    response = requests.post(url, headers=headers, json=data)
    result = response.json()
    return result['output']['choices'][0]['text'].strip()

def generate_message_with_submit(input_text, model):
    return generate_message(input_text, model)

iface = gr.Interface(
    fn=generate_message_with_submit,
    inputs=[
        gr.TextArea(lines=20, label="Enter Prompt"),
        gr.Radio(['meta-llama/Llama-2-70b-chat-hf', 'iamplus/model1-70b'], label="Model",value='meta-llama/Llama-2-70b-chat-hf')
    ],
    outputs=gr.TextArea(label="AI Assistant Message"),
    title="AI Assistant Message Generator",
    description="Generate an appropriate message for an AI assistant to let the user know that it's working on a task and will get back soon.",
    allow_flagging=False
)

iface.launch(share=True)