app.py

#!/usr/bin/env python
# or gradio app.py

import gradio as gr

from chat import iface_chat
from prompt import iface_prompt

with gr.Blocks() as iface:
    gr.Markdown(
        """# Petals playground
        **Let's play with prompts and inference settings for various big LLM models running on Petals!**

        This space uses websocket API of [chat.petals.dev](http://chat.petals.dev). Health status of Petals network [lives here](http://health.petals.dev)."""
    )

    gr.TabbedInterface([iface_prompt, iface_chat], ["Prompt mode", "Chat mode"])

# Queues are required to enable generators
iface.queue(concurrency_count=5, max_size=50)
iface.launch(show_error=True)