app.py

import gradio as gr
from huggingface_hub import InferenceClient
from llama_cpp import Llama

# Initialize the InferenceClient
client = InferenceClient()

llm = Llama.from_pretrained(
    repo_id="bartowski/Ministral-8B-Instruct-2410-GGUF",
    filename="Ministral-8B-Instruct-2410-Q4_K_M.gguf",
)

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    response = ""

    # Use the client to get the chat completion
    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message['choices'][0]['delta']['content']
        response += token
        yield response

demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly, conversational, helpful, and informative chatbot, designed to help users as best as possible. Responses should be quirky and fun to read, including the use of appropriate emojis in answers, wherever necesssary.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, 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 (nucleus sampling)",
        ),
    ],
)

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