app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline

model_id = "ibm-granite/granite-3.1-2b-instruct"  # Hugging Face model ID
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")

pipe = pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    max_length=512,
    temperature=0.7
)

def generate_answer(prompt):
    result = pipe(prompt)[0]["generated_text"]
    return result

demo = gr.Interface(
    fn=generate_answer,
    inputs="text",
    outputs="text",
    title="Granite 3.1 2B Instruct - RAG Answering"
)

demo.launch()