app.py

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
import gradio as gr
import os

hf_token = os.environ.get('HUGGINGFACE_TOKEN')

# Define the device
device = "cuda" if torch.cuda.is_available() else "cpu"

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained('stabilityai/stablelm-zephyr-3b', token=hf_token)
model = AutoModelForCausalLM.from_pretrained(
    'stabilityai/stablelm-zephyr-3b',
    trust_remote_code=True,
    device_map="auto",
    token=hf_token
)
model.to(device)

class ChatBot:
    def __init__(self):
        self.history = []

    def predict(self, user_input, system_prompt="You are an expert medical analyst:"):
        prompt = [{'role': 'user', 'content': user_input}, {'role': 'system', 'content': system_prompt}]
        inputs = tokenizer.apply_chat_template(
            prompt,
            add_generation_prompt=True,
            return_tensors='pt'
        )

        # Generate a response using the model
        tokens = model.generate(
            inputs.to(model.device),
            max_new_tokens=250,
            temperature=0.8,
            do_sample=False
        )

        # Decode and return the response
        response_text = tokenizer.decode(tokens[0], skip_special_tokens=False)
        return response_text
        
        # Free up memory
        del tokens
        torch.cuda.empty_cache()
        return response_text

bot = ChatBot()

title = "👋🏻Welcome to 🌟Tonic's🗽Stable🌟LM 3B🚀Chat"
description = """
You can use this Space to test out the current model [stabilityai/stablelm-zephyr-3b](https://huggingface.co/stabilityai/stablelm-zephyr-3b)
You can also use 😷StableMed⚕️ on your laptop & by cloning this space. 🧬🔬🔍 Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/Tonic/TonicsStableLM3B?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3> 
Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's🛠️community on 👻Discord: [Discord](https://discord.gg/GWpVpekp) On 🤗Huggingface: [TeamTonic](https://huggingface.co/TeamTonic) & [MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Polytonic](https://github.com/tonic-ai) & contribute to 🌟 [PolyGPT](https://github.com/tonic-ai/polygpt-alpha)
"""
examples = [["What is the proper treatment for buccal herpes?", "Please provide information on the most effective antiviral medications and home remedies for treating buccal herpes."]]

iface = gr.Interface(
    fn=bot.predict,
    title=title,
    description=description,
    examples=examples,
    inputs=["text", "text"],  # Take user input and system prompt separately
    outputs="text",
    theme="ParityError/Anime"
)

iface.launch()