app.py

import gradio as gr
from llama_cpp import Llama

model = "Qwen/Qwen2-7B-Instruct-GGUF"
llm = Llama.from_pretrained(
    repo_id=model,
    filename="qwen2-7b-instruct-q4_k_m.gguf",
    verbose=False,
    use_mmap=False,
    use_mlock=True,
    n_threads=2,
    n_threads_batch=2,
    n_ctx=40000,
)


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 = llm.create_chat_completion(
        messages=messages,
        max_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
    )
    return response["choices"][0]["message"]["content"]


demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(
            value="You are a helpful assistant.",
            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)",
        ),
    ],
    description=model,
)


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