app.py

import gradio as gr
import os
import time
from cerebras.cloud.sdk import Cerebras

# Set up the Cerebras client
client = Cerebras(api_key=os.getenv("CEREBRAS_API_KEY"))

def chat_with_cerebras(user_input):
    """
    Handles interaction with the Cerebras model.
    Sends user input and returns the model's response along with compute time.
    """
    # Start compute time measurement
    start_time = time.time()

    try:
        # Create a chat stream with Cerebras
        stream = client.chat.completions.create(
            messages=[
                {"role": "system", "content": "You are a helpful assistant."},
                {"role": "user", "content": user_input}
            ],
            model="llama-3.3-70b",
            stream=True,
            max_completion_tokens=1024,
            temperature=0.2,
            top_p=1
        )

        # Collect response from the stream
        response = ""
        for chunk in stream:
            response += chunk.choices[0].delta.content or ""

        # End compute time measurement
        compute_time = time.time() - start_time

        return response, f"Compute Time: {compute_time:.2f} seconds"

    except Exception as e:
        return "Error: Unable to process your request.", str(e)

# Gradio interface
def gradio_ui():
    with gr.Blocks() as demo:
        gr.Markdown("""# Cerebras AI Chatbot\nChat with a state-of-the-art AI model.""")

        with gr.Row():
            user_input = gr.Textbox(label="Your Message", placeholder="Type your message here...")
            response = gr.Textbox(label="AI Response", interactive=False)
            compute_time = gr.Textbox(label="Compute Time", interactive=False)

        submit_button = gr.Button("Submit")

        # Define interaction logic
        submit_button.click(chat_with_cerebras, inputs=user_input, outputs=[response, compute_time])

    return demo

# Run the Gradio app
demo = gradio_ui()
demo.launch()