app.py

import gradio as gr
from huggingface_hub import InferenceClient
import asyncio

# Define the model ID
model_id = 'Aksh1t/mistral-7b-oig-unsloth-merged'

# Initialize the Hugging Face inference client
client = InferenceClient()

async def generate_text(prompt):
    # Use the Hugging Face client to generate text asynchronously
    async with client.chat(model_id=model_id) as chat:
        response = await chat(prompt)
        return response['generated_text']

# Create Gradio interface
inputs = gr.Textbox(label="Enter a prompt", lines=2)
outputs = gr.Textbox(label="Generated Text", readonly=True)

def predict(prompt):
    output_text = asyncio.run(generate_text(prompt))
    return output_text

iface = gr.Interface(fn=predict, inputs=inputs, outputs=outputs, title="Hugging Face Text Generation Model", description=f"Generate text based on a prompt using model '{model_id}'")

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