# app.py
import gradio as gr
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Initialize model and tokenizer
MODEL_NAME = "kaiiddo/A3ON"
TOKEN = "YOUR_HF_TOKEN"  # Set in HF Secrets

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained(
    MODEL_NAME,
    token=TOKEN,
    trust_remote_code=True
)

model = AutoModelForCausalLM.from_pretrained(
    MODEL_NAME,
    token=TOKEN,
    torch_dtype=torch.float16,
    low_cpu_mem_usage=True,
    trust_remote_code=True
)

def generate_text(prompt, max_new_tokens=200, temperature=0.9, top_p=0.9):
    """Generate text using the A3ON model"""
    inputs = tokenizer.encode(prompt, return_tensors="pt")
    
    with torch.no_grad():
        outputs = model.generate(
            inputs,
            max_new_tokens=max_new_tokens,
            temperature=temperature,
            top_p=top_p,
            do_sample=True,
            pad_token_id=tokenizer.eos_token_id
        )
    
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

# Gradio interface
with gr.Blocks(title="A3ON Text Generator") as demo:
    gr.Markdown("# A3ON Text Generator")
    gr.Markdown("Generate text using the A3ON model. Adjust parameters for creative outputs.")
    
    with gr.Row():
        with gr.Column():
            prompt = gr.Textbox(
                label="Input Prompt",
                placeholder="Enter your prompt here...",
                lines=5
            )
            max_tokens = gr.Slider(
                50, 500, value=200, label="Max New Tokens"
            )
            temp = gr.Slider(
                0.1, 2.0, value=0.9, label="Temperature"
            )
            top_p = gr.Slider(
                0.1, 1.0, value=0.9, label="Top-P (Nucleus Sampling)"
            )
            generate_btn = gr.Button("Generate")
        
        with gr.Column():
            output = gr.Textbox(
                label="Generated Text",
                lines=10,
                interactive=False
            )
    
    generate_btn.click(
        generate_text,
        inputs=[prompt, max_tokens, temp, top_p],
        outputs=output
    )
    
    gr.Examples(
        examples=[
            ["Once upon a time in a galaxy far far away"],
            ["The secret to happiness is"],
            ["In the year 2050, artificial intelligence"]
        ],
        inputs=[prompt]
    )

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