app.py

import gradio as gr
import torch
from transformers import pipeline

def generate_text(prompt):
    messages = [
        {"role": "system", "content": "You are a code assistant"},
        {"role": "user", "content": prompt},
    ]
    formatted_prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
    outputs = pipe(formatted_prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
    generated_text = outputs[0]["generated_text"]
    return generated_text

# Load the model pipeline
pipe = pipeline("text-generation", model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", torch_dtype=torch.bfloat16, device_map="auto")

# Create Gradio interface
iface = gr.Interface(fn=generate_text, inputs="text", outputs="text", live=True, title="Chatbot Assistant")
iface.launch()