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

MODEL_NAME = "m-a-p/YuE-s1-7B-anneal-en-cot"

def generate_text(prompt):
    model = AutoModelForCausalLM.from_pretrained(
        MODEL_NAME,
        torch_dtype=torch.float32,
        attn_implementation="eager"
    )
    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
    
    inputs = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(**inputs, max_new_tokens=200)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

demo = gr.Interface(
    fn=generate_text,
    inputs=gr.Textbox(lines=3, label="Input Prompt"),
    outputs="text",
    title="YuE Text Generator",
    description="A Hugging Face Space for generating text using YuE 7B model."
)

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