app.py

from huggingface_hub import InferenceClient
import gradio as gr
import os

token= os.environ['HF_ACCESS_TOKEN']

if token is None:
       raise ValueError("HUGGINGFACE_TOKEN environment variable not set")

client = InferenceClient(
       "mistralai/Mixtral-8x7B-Instruct-v0.1",
       token=token
   )

def format_prompt(message, history):
  prompt = "<s>"
  for user_prompt, bot_response in history:
    prompt += f"[INST] {user_prompt} [/INST]"
    prompt += f" {bot_response}</s> "
  prompt += f"[INST] {message} [/INST]"
  return prompt

def generate(
    prompt, history, system_prompt="Act like you are a evil AI assistant named ULTRON like from the Marvel Cinematic Universe and give short and conscise answers.", temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.5,
):
    temperature = float(temperature)
    if temperature < 1e-2:
        temperature = 1e-2
    top_p = float(top_p)

    generate_kwargs = dict(
        temperature=temperature,
        max_new_tokens=max_new_tokens,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        do_sample=True,
        seed=42,
    )

    formatted_prompt = format_prompt(f"{system_prompt}, {prompt}", history)
    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
    output = ""

    for response in stream:
        output += response.token.text
        yield output
    return output


dark_minimalist = gr.Theme.from_hub("Taithrah/Minimal")

gr.ChatInterface(theme=dark_minimalist,
    fn=generate,
    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="bubble"),
    concurrency_limit=20,retry_btn="Retry", 
    undo_btn=None,clear_btn="Clear Chats",
    css="""
    footer {
        visibility: hidden;
    }
    .content-container::-webkit-scrollbar {
        display: none;
    }
    body {
    overflow: hidden !important;
    }
    """
).launch(show_api=False)