app.py

#!/usr/bin/env python

import os
from threading import Thread
from typing import Iterator

import gradio as gr
import spaces
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer

DESCRIPTION = "# Mistral-7B"

if not torch.cuda.is_available():
    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"

MAX_MAX_NEW_TOKENS = 2048
DEFAULT_MAX_NEW_TOKENS = 1024
MAX_INPUT_TOKEN_LENGTH = 4096

if torch.cuda.is_available():
    model_id = "codys12/MergeLlama-7b"
    model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map=0, cache_dir="/data")
    model.cuda()
    tokenizer = AutoTokenizer.from_pretrained(model_id)


@spaces.GPU
def generate(
    message: str,
    chat_history: list[tuple[str, str]],
    max_new_tokens: int = 1024,
    temperature: float = 0.6,
    top_p: float = 0.9,
    top_k: int = 50,
    repetition_penalty: float = 1.2,
) -> Iterator[str]:
    conversation = []
    current_input = ""
    for user, assistant in chat_history:
        current_input += user
        current_input += assistant

    current_input += message
    
    device = "cuda:0"
    input_ids = tokenizer(current_input, return_tensors="pt").to(device)

    if len(input_ids) > MAX_INPUT_TOKEN_LENGTH:
        input_ids = input_ids[-MAX_INPUT_TOKEN_LENGTH:]
        gr.Warning("Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")

    input_ids = tokenizer(current_input, return_tensors="pt").to("cuda")

    # Generate 
    output_ids = model.generate(input_ids=input_ids,
                              max_new_tokens=max_new_tokens, 
                              do_sample=True,
                              top_p=top_p,
                              top_k=top_k,
                              temperature=temperature,
                              repetition_penalty=repetition_penalty)

    # Stream output    
    for id in output_ids.tolist()[0]:
     yield tokenizer.decode(id)


chat_interface = gr.ChatInterface(
    fn=generate,
    additional_inputs=[
        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=0.6,
        ),
        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=[
        ["Hello there! How are you doing?"],
        ["Can you explain briefly to me what is the Python programming language?"],
        ["Explain the plot of Cinderella in a sentence."],
        ["How many hours does it take a man to eat a Helicopter?"],
        ["Write a 100-word article on 'Benefits of Open-Source in AI research'"],
    ],
)

with gr.Blocks(css="style.css") as demo:
    gr.Markdown(DESCRIPTION)
    gr.DuplicateButton(
        value="Duplicate Space for private use",
        elem_id="duplicate-button",
        visible=os.getenv("SHOW_DUPLICATE_BUTTON") == "1",
    )
    chat_interface.render()

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