app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("apple/DCLM-Baseline-7B-8k")
model = AutoModelForCausalLM.from_pretrained("apple/DCLM-Baseline-7B-8k")

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})

    prompt = "".join([f"{'[|Human|] ' if msg['role'] == 'user' else '[|AI|] '}{msg['content']}" for msg in messages])

    inputs = tokenizer(prompt, return_tensors="pt")
    gen_kwargs = {
        "max_new_tokens": max_tokens,
        "top_p": top_p,
        "temperature": temperature,
        "do_sample": True,
        "repetition_penalty": 1.1
    }
    with torch.no_grad():
        output = model.generate(inputs['input_ids'], **gen_kwargs)
    response = tokenizer.decode(output[0].tolist(), skip_special_tokens=True)[len(prompt):]

    yield response

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=0.7, 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()