app.py

import gradio as gr
import requests
import os

# Hugging Face Inference API for Qwen3-Coder-Next (Mac model)
HF_MODEL = "Qwen/Qwen3-Coder-Next"
HF_TOKEN = os.getenv("HF_TOKEN")  # Set this in Space Secrets

HEADERS = {"Authorization": f"Bearer {HF_TOKEN}"} if HF_TOKEN else {}

def generate_response(prompt):
    """
    Send the prompt to the Hugging Face Inference API and return the response.
    """
    payload = {"inputs": prompt}
    try:
        response = requests.post(
            f"https://api-inference.huggingface.co/models/{HF_MODEL}",
            headers=HEADERS,
            json=payload,
            timeout=60  # prevent infinite wait
        )
        response.raise_for_status()
        result = response.json()
        # Check if the API returned generated text
        if isinstance(result, list) and "generated_text" in result[0]:
            return result[0]["generated_text"]
        return str(result)
    except Exception as e:
        return f"Error: {e}"

# Gradio UI
with gr.Blocks() as demo:
    gr.Markdown("# Qwen3-Coder-Next — Coding Assistant")
    user_input = gr.Textbox(label="Prompt", placeholder="Type your code request here...")
    output = gr.Textbox(label="Response")
    user_input.submit(generate_response, inputs=user_input, outputs=output)
    gr.Button("Submit").click(generate_response, inputs=user_input, outputs=output)

demo.launch(server_name="0.0.0.0", server_port=8080)