app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "exzort/VineBot"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

def chatbot(message):
    prompt = f"Q: {message}\nA:"
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=200, pad_token_id=tokenizer.eos_token_id)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)

    # remove the prompt part so only the answer remains
    if "A:" in response:
        response = response.split("A:")[-1].strip()
    return response


demo = gr.Interface(
    fn=chatbot,
    inputs="text",
    outputs="text",
    title="VineBot Chat",
    description="Ask me anything about the ministry manual!"
)

if __name__ == "__main__":
    demo.launch()