app.py

import spaces
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
from threading import Thread

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
model_name = "Merdeka-LLM/merdeka-llm-3.2b-128k-instruct"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

streamer = TextIteratorStreamer(tokenizer, timeout=300, skip_prompt=True, skip_special_tokens=True)

@spaces.GPU
def respond(
    message,
    history: list[tuple[str, str]],
    # system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [
        {"role": "system", "content": "You are a professional lawyer who is familiar with Malaysia Law."}
    ]

    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 = ""

    text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True,
    )
    model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

    generate_kwargs = dict(
        model_inputs,
        max_new_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p,
        streamer=streamer
    )
    t = Thread(target=model.generate, kwargs=generate_kwargs)
    t.start()
    for new_token in streamer:
      if new_token != '<':
          response += new_token
          yield response

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        # gr.Textbox(value="You are a friendly Chatbot.", 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=0.1, 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(
        
    )