app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import spaces
device = "cuda"

tokenizer = AutoTokenizer.from_pretrained("NoaiGPT/777")
model = AutoModelForSeq2SeqLM.from_pretrained("NoaiGPT/777").to(device)

@spaces.GPU
def generate_title(text):
    input_ids = tokenizer(f'paraphraser: {text}', return_tensors="pt", padding="longest", truncation=True, max_length=64).input_ids.to(device)
    outputs = model.generate(
        input_ids,
        num_beams=4,
        num_beam_groups=4,
        num_return_sequences=4,
        repetition_penalty=10.0,
        diversity_penalty=3.0,
        no_repeat_ngram_size=2,
        temperature=0.8,
        max_length=64
    )
    return tokenizer.batch_decode(outputs, skip_special_tokens=True)

def gradio_generate_title(text):
    titles = generate_title(text)
    return "\n\n".join(titles)

iface = gr.Interface(
    fn=gradio_generate_title,
    inputs=gr.Textbox(lines=5, label="Input Text"),
    outputs=gr.Textbox(lines=10, label="Generated Titles"),
    title="Title Generator",
    description="Generate multiple paraphrased titles from input text using NoaiGPT/777 model."
)

iface.launch()