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

# Model loading with memory optimization
cache_dir = "./model_cache"
os.makedirs(cache_dir, exist_ok=True)

model_name = "Qwen/Qwen1.5-0.5B"

tokenizer = AutoTokenizer.from_pretrained(
    model_name,
    cache_dir=cache_dir,
    trust_remote_code=True
)

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    cache_dir=cache_dir,
    trust_remote_code=True
).to("cuda" if torch.cuda.is_available() else "cpu")

# Generation configuration
generation_config = GenerationConfig.from_pretrained(model_name)

def generate_text(prompt, temperature, max_new_tokens):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    
    with torch.inference_mode():
        outputs = model.generate(
            **inputs,
            max_new_tokens=int(max_new_tokens),
            temperature=float(temperature),
            pad_token_id=tokenizer.eos_token_id
        )
    
    response = tokenizer.decode(outputs[0][len(inputs["input_ids"][0]):], skip_special_tokens=True)
    return response.strip()

# Gradio interface
with gr.Blocks(theme="soft", title="Qwen Text Generation") as demo:
    gr.Markdown("# 🧠 Qwen1.5-0.5B Text Generation")
    
    with gr.Row():
        with gr.Column():
            prompt = gr.Textbox(
                label="Input Prompt",
                placeholder="Enter your instruction or question...",
                lines=5
            )
            temperature = gr.Slider(
                minimum=0.1, maximum=2.0, value=0.7, step=0.1,
                label="Creativity (Temperature)"
            )
            max_new_tokens = gr.Slider(
                minimum=50, maximum=1000, value=200, step=50,
                label="Max New Tokens"
            )
            generate_btn = gr.Button("✨ Generate", variant="primary")
        
        with gr.Column():
            output = gr.Textbox(label="Model Response", lines=10, interactive=False)
    
    generate_btn.click(
        fn=generate_text,
        inputs=[prompt, temperature, max_new_tokens],
        outputs=output
    )
    
    gr.Markdown("""
    ### ℹ️ Tips
    - Higher **temperature** = more creative/chaotic responses
    - Lower **temperature** = more deterministic answers
    - Adjust **max tokens** for longer/shorter outputs
    """)

demo.launch()