app.py

# Gradio Chatbot with Gemma-3-1B-IT

from gradio import Interface, Chatbot
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model_name = 'google/gemma-3-1b-it'
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Chatbot handler
def respond(message, chat_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

# Create interface
chatbot = Chatbot()
iface = Interface(fn=respond, inputs=chatbot, outputs=chatbot)

# Launch app
if __name__ == '__main__':
    iface.launch()