app.py

import os
from threading import Thread
from typing import Iterator
import os
from huggingface_hub import login
import gradio as gr
import spaces
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer

MAX_MAX_NEW_TOKENS = 128
DEFAULT_MAX_NEW_TOKENS = 1024
MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
model = None
tokenizer = None

def load_model():
    global model, tokenizer
    model_id = "stabilityai/ar-stablelm-2-chat"
    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto", trust_remote_code=True)
    tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
    model.generation_config.pad_token_id = model.generation_config.eos_token_id



def generate(
    message: str,
    chat_history: list[dict],
    system_prompt: str = "",
    max_new_tokens: int = 128,
    temperature: float = 0.6,
    top_p: float = 0.9,
    top_k: int = 50,
    repetition_penalty: float = 1.2,
) -> Iterator[str]:
    conversation = []
    if system_prompt:
        conversation.append({"role": "system", "content": system_prompt})
    conversation += chat_history
    conversation.append({"role": "user", "content": message})

    input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt")
    if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
        input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
        gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
    input_ids = input_ids.to(model.device)

    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
    generate_kwargs = dict(
        {"input_ids": input_ids},
        streamer=streamer,
        max_new_tokens=max_new_tokens,
        do_sample=True,
        eos_token_id=tokenizer.eos_token_id,  # Stop generation at <EOS>
        temperature=temperature,
        top_p=top_p,
        top_k=top_k
    )
    t = Thread(target=model.generate, kwargs=generate_kwargs)
    t.start()

    outputs = []
    for text in streamer:
        outputs.append(text)
        yield "".join(outputs)


chat_interface = gr.ChatInterface(
    fn=generate,
    additional_inputs=[
        gr.Textbox(label="System prompt", lines=6),
        gr.Slider(
            label="Max new tokens",
            minimum=1,
            maximum=MAX_MAX_NEW_TOKENS,
            step=1,
            value=DEFAULT_MAX_NEW_TOKENS,
        ),
        gr.Slider(
            label="Temperature",
            minimum=0.1,
            maximum=4.0,
            step=0.1,
            value=0.7,
        ),
        gr.Slider(
            label="Top-p (nucleus sampling)",
            minimum=0.05,
            maximum=1.0,
            step=0.05,
            value=0.9,
        ),
        gr.Slider(
            label="Top-k",
            minimum=1,
            maximum=1000,
            step=1,
            value=50,
        ),
        gr.Slider(
            label="Repetition penalty",
            minimum=1.0,
            maximum=2.0,
            step=0.05,
            value=1.2,
        ),
    ],
    stop_btn=None,
    examples=[
        ["السلام عليكم"],
        ["اعرب الجملة التالية: ذهبت الى السوق"]
    ],
    cache_examples=False,
    type="messages",
)

with gr.Blocks(css_paths="style.css", fill_height=True) as demo:
    def authenticate_token(token):
        try:
            login(token = token)
            load_model()
            return "Authenticated successfully"
        except:
            return "Invalid token. Please try again."

    # Components
    token_input = gr.Textbox(label="Hugging Face Access Token", type="password", placeholder="Enter your token here...")
    auth_button = gr.Button("Authenticate")
    output = gr.Textbox(label="Output")
    auth_button.click(fn=authenticate_token, inputs=token_input, outputs=output)
    chat_interface.render()

    

if __name__ == "__main__":
    demo.queue(max_size=20).launch()