app.py

import subprocess
subprocess.check_call(["pip", "install", "-q", "gradio", "transformers", "python-dotenv", "torch"])
#!pip install -q gradio
#!pip install gradio
#!pip install --quiet gradio
#!pip install transformers
#!pip install python-dotenv

import gradio as gr
from transformers import GPTNeoForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neo-2.7B")
model = GPTNeoForCausalLM.from_pretrained("EleutherAI/gpt-neo-2.7B")

def generate_text(prompt):
    input_ids = tokenizer.encode(prompt, return_tensors="pt")
    output = model.generate(input_ids, max_length=1024, temperature=0.5, do_sample=True)
    return tokenizer.decode(output[0], skip_special_tokens=True)

def chatbot(input, history=[]):
    output = generate_text(input)
    history.append((input, output))
    return history, history

gr.Interface(
    fn=chatbot,
    inputs=["text", "state"],
    outputs=["chatbot", "state"]
).launch(debug=True)