app.py

from huggingface_hub import InferenceClient
import gradio as gr

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

def format_prompt(message, history):
    global system_prompt_sent
    prompt = "<s>"

    if not any(f"[INST] {system_prompt} [/INST]" in user_prompt for user_prompt, _ in history):
        prompt += f"[INST] {system_prompt} [/INST]"
        system_prompt_sent = True

    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=0.9, max_new_tokens=4096, top_p=0.95, repetition_penalty=1.0,
):
    global system_prompt_sent
    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(prompt, history)
    
    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=True)
    output = ""

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

    return output

chat_interface = gr.ChatInterface(
    fn=generate,
    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=False, layout="vertical", height=700),
    concurrency_limit=9,
    theme="soft",
    submit_btn="Enviar",
)

chat_interface.launch(show_api=False)