app.py

import gradio as gr

from transformers import pipeline
import torch

pipe = pipeline(
    "image-text-to-text",
    model="google/gemma-3-4b-it",
    device="cuda",
    torch_dtype=torch.bfloat16
)

def greet(name):
    messages = [
    {
        "role": "system",
        "content": [{"type": "text", "text": "You are a helpful assistant."}]
    },
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": name}
        ]
    }
    ]

    output = pipe(text=messages, max_new_tokens=200)
    return output

demo = gr.Interface(fn=greet, inputs="text", outputs="text")
demo.launch()