app.py

import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer

# LLM model
model_name = "swap-uniba/LLaMAntino-2-7b-hf-ITA"  # Sostituisci con il modello desiderato
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto")

# Funzione di inferenza
def chat(input_text):
    inputs = tokenizer(input_text, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=150)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Interfaccia Gradio
iface = gr.Interface(fn=chat, inputs="text", outputs="text")
iface.launch()