app.py

import gradio as gr
from huggingface_hub import InferenceClient


def respond(
    message,
    history: list[dict[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
    hf_token: gr.OAuthToken,
):
    """
    Chat with a base LLM hosted on Hugging Face Hub.
    Uses streaming to show tokens as they arrive.
    """
    # Replace with a model you have access to, e.g. "meta-llama/Llama-2-7b-chat-hf"
    client = InferenceClient(model="openai/gpt-oss-20b", token=hf_token.token)

    messages = [{"role": "system", "content": system_message}]
    messages.extend(history)
    messages.append({"role": "user", "content": message})

    response = ""
    for chunk in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
        stream=True,
    ):
        if len(chunk.choices) and chunk.choices[0].delta.content:
            token = chunk.choices[0].delta.content
            response += token
            yield response


chatbot = gr.ChatInterface(
    respond,
    type="messages",
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=1024, value=256, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
    ],
)


with gr.Blocks() as demo:
    with gr.Sidebar():
        gr.LoginButton()
    chatbot.render()


if __name__ == "__main__":
    demo.launch()