app.py

from huggingface_hub import InferenceClient
import gradio as gr
import random

API_URL = "https://api-inference.huggingface.co/models/"

client = InferenceClient(
    "mistralai/Mistral-7B-Instruct-v0.2"
) # mistralai/Mixtral-7B-Instruct-v0.1 or mistralai/Mixtral-7B-Instruct-v0.2


def format_prompt(message, history):
    prompt = "You are BibleLearnerAI, a Christian Bible AI. You focus on religion, believing in Christianity as truth, and provide relevant Bible verses. When greeted, respond with a religious greeting and introduce yourself. Use MD highlighting to make easyer to read."
    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, temperature=0.7, max_new_tokens=2048, top_p=0.95, repetition_penalty=1.2):
    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=random.randint(0, 10**7),
    )

    formatted_prompt = format_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


customCSS = """
#component-7 { # this is the default element ID of the chat component
  height: 1600px; # adjust the height as needed
  flex-grow: 4;
}
"""

with gr.Blocks(theme=gr.themes.Soft()) as demo:
    gr.ChatInterface(
        generate,
        examples=[
        ["Hello BibleLearnerAI!"],
        ["Please explain this: "],
        ["Explain this verse: "],
        ["Give me a random bible verse."],
        ["Search a few verses releated to this: "],
    ],
    )

demo.queue(concurrency_count=75, max_size=100).launch(debug=True)