app.py

import os
import logging
import gradio as gr
from huggingface_hub import InferenceClient

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s | %(levelname)s | %(message)s",
    datefmt="%Y-%m-%d %H:%M:%S",
)
logger = logging.getLogger(__name__)

# Environment variables for configuration
HF_TOKEN = os.environ.get("HF_TOKEN", "")
MODEL_ID = os.environ.get("MODEL_ID", "Qwen/Qwen2.5-72B-Instruct")
DEFAULT_MAX_TOKENS = int(os.environ.get("DEFAULT_MAX_TOKENS", "100"))
DEFAULT_TEMPERATURE = float(os.environ.get("DEFAULT_TEMPERATURE", "0.7"))

logger.info(f"HF_TOKEN configured: {bool(HF_TOKEN)}")
logger.info(f"MODEL_ID: {MODEL_ID}")
logger.info(f"DEFAULT_MAX_TOKENS: {DEFAULT_MAX_TOKENS}")
logger.info(f"DEFAULT_TEMPERATURE: {DEFAULT_TEMPERATURE}")

client = InferenceClient(token=HF_TOKEN) if HF_TOKEN else InferenceClient()
logger.info("InferenceClient initialized")



def generate(prompt: str, max_tokens: int, temperature: float) -> str:
    """Generate text from prompt."""
    logger.info(f"generate() called | prompt_len={len(prompt)} | max_tokens={max_tokens} | temp={temperature}")
    
    if not prompt.strip():
        logger.warning("Empty prompt received")
        return "✏️ Enter a prompt!"
    
    try:
        logger.info(f"Calling chat.completions | model={MODEL_ID}")
        response = client.chat.completions.create(
            model=MODEL_ID,
            messages=[{"role": "user", "content": prompt}],
            max_tokens=max_tokens,
            temperature=temperature,
        )
        output = response.choices[0].message.content
        logger.info(f"Generated {len(output)} chars")
        return output
    except Exception as e:
        logger.error(f"API error: {e}")
        return f"❌ Error: {e}"


logger.info("Building Gradio interface...")

with gr.Blocks(title="Text Playground") as demo:
    gr.Markdown("# ✍️ Text Playground\nGenerate creative text with powerful language models!")

    prompt = gr.Textbox(
        label="Your prompt",
        placeholder="Write a haiku about coding...",
        lines=3,
        autofocus=True,
    )

    with gr.Row(equal_height=True):
        max_tokens = gr.Slider(
            minimum=10, maximum=500, value=DEFAULT_MAX_TOKENS, step=10,
            label="Max tokens"
        )
        temperature = gr.Slider(
            minimum=0.1, maximum=1.5, value=DEFAULT_TEMPERATURE, step=0.1,
            label="Temperature (creativity)"
        )

    output = gr.Textbox(label="Generated text", lines=6, interactive=False)
    btn = gr.Button("Generate ✨", variant="primary")

    btn.click(generate, inputs=[prompt, max_tokens, temperature], outputs=output)
    prompt.submit(generate, inputs=[prompt, max_tokens, temperature], outputs=output)

    gr.Examples(
        examples=[
            ["Write a haiku about machine learning", 50, 0.8],
            ["Explain quantum computing to a 5-year-old", 150, 0.5],
            ["Give me 3 creative startup ideas", 150, 0.9],
        ],
        inputs=[prompt, max_tokens, temperature],
    )

demo.queue()
logger.info("Starting Gradio server...")
demo.launch()