app.py

import streamlit as st
from huggingface_hub import InferenceClient

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

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, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
    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

# Create text input for user message
message_input = st.text_input("You:", "")

# Create text input for system prompt
system_prompt_input = st.text_input("System Prompt:", "You are a helpful assistant.")

# Create sliders for temperature, max new tokens, top-p, and repetition penalty
temperature_slider = st.slider("Temperature", 0.0, 1.0, 0.9)
max_new_tokens_slider = st.slider("Max new tokens", 0, 1048, 256)
top_p_slider = st.slider("Top-p (nucleus sampling)", 0.0, 1.0, 0.95)
repetition_penalty_slider = st.slider("Repetition penalty", 1.0, 2.0, 1.0)

# Create button to generate response
if st.button("Generate"):
    # Create empty list to store conversation history
    history = []

    # Call generate function with user message, system prompt, and slider values
    output = generate(message_input, history, system_prompt_input, temperature=temperature_slider, max_new_tokens=max_new_tokens_slider, top_p=top_p_slider, repetition_penalty=repetition_penalty_slider)

    # Display generated response
    st.write("Assistant:", output)

    # Add user message and generated response to conversation history
    history.append((message_input, output))