app.py

import gradio as gr
import requests

# Function to send your prompt to NVIDIA LLaMA 4 Scout
def talk_to_llama(prompt):
    url = "https://api.nvcf.nvidia.com/v2/completions"
    headers = {
        "Authorization": "Bearer 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",
        "Content-Type": "application/json"
    }
    data = {
        "messages": [{"role": "user", "content": prompt}]
    }
    response = requests.post(url, headers=headers, json=data)
    try:
        return response.json()["choices"][0]["message"]["content"]
    except:
        return "Something went wrong. Here's what the server said: " + str(response.text)

# Build the chatbot interface
chat = gr.Interface(
    fn=talk_to_llama,
    inputs="text",
    outputs="text",
    title="Chat with LLaMA 4 Scout",
    description="Ask anything! This chatbot uses NVIDIA’s 3.5M token LLaMA 4 Scout model."
)

chat.launch()