app.py

import subprocess
import gradio as gr

# Compile the model using "!make runfast"
try:
    subprocess.run(["make", "runfast"], check=True, shell=True)
    print("Model compilation successful.")
except subprocess.CalledProcessError as e:
    print("Error:", e)
    print(e.stderr)

def chatbot(prompt, temperature, topt, maxtoken, model_file):
    command = ["./run", model_file, "-t", str(temperature), "-p", str(topt), "-n", str(maxtoken), "-i", f"{prompt}"]
    try:
        result = subprocess.run(command, capture_output=True, text=True, check=True, shell=False)
        response = result.stdout
    except subprocess.CalledProcessError as e:
        response = "Error occurred while processing the request."
    return response

with gr.Blocks() as demo:
    gr.Markdown("HF Spaces for TinyStories")
    with gr.Row():
        with gr.Column():
            inp = gr.Textbox(placeholder="Type the beginning of the story")
            with gr.Row():
                with gr.Column():
                    temperature_slider = gr.Slider(minimum=0.1, maximum=2.0, value=0.8, label="Temperature")
                    topt_slider = gr.Slider(minimum=0.1, maximum=1.0, value=0.9, label="Topt")
                    maxtoken_slider = gr.Slider(minimum=16, maximum=512, value=256, label="Max Tokens")
                model_file_dropdown = gr.Dropdown(choices=["stories15M.bin", "stories42M.bin", "stories110M.bin"], label="Model File", value="stories15M.bin")
        out = gr.Textbox()
    btn = gr.Button("Run")
    
    btn.click(fn=chatbot, inputs=[inp, temperature_slider, topt_slider, maxtoken_slider, model_file_dropdown], outputs=out)

demo.launch()