app.py

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

# Load model & tokenizer
MODEL_NAME = "ubiodee/Test_Plutus"

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)
model.eval()

if torch.cuda.is_available():
    model.to("cuda")

# Response function
def generate_response(prompt):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

    with torch.no_grad():
        outputs = model.generate(
            **inputs,
            max_new_tokens=200,
            temperature=0.7,
            top_p=0.9,
            do_sample=True,
            eos_token_id=tokenizer.eos_token_id,
            pad_token_id=tokenizer.pad_token_id,
        )
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)

    # Remove the prompt from the output to return only the answer
    if response.startswith(prompt):
        response = response[len(prompt):].strip()

    return response

# Gradio UI
demo = gr.Interface(
    fn=generate_response,
    inputs=gr.Textbox(label="Enter your prompt", lines=4, placeholder="Ask about Plutus..."),
    outputs=gr.Textbox(label="Model Response"),
    title="Cardano Plutus AI Assistant",
    description="Write Plutus smart contracts on Cardano blockchain."
)

demo.launch()