import gradio as gr
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

MODEL_NAME = "google/flan-t5-base"

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME)

def generate_response(message, history):
    inputs = tokenizer(message, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=100)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return response

gr.ChatInterface(generate_response).launch()