app.py

from huggingface_hub import InferenceClient
import gradio as gr

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, temperature=0.2, max_new_tokens=30000, top_p=0.9, repetition_penalty=1.0,
):
    temperature = max(float(temperature), 0.01)
    top_p = max(min(float(top_p), 1.0), 0.0)
    repetition_penalty = max(float(repetition_penalty), 0.01)

    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)

    # Generate text
    response = client.text_generation(formatted_prompt, **generate_kwargs)
    generated_text = response["generated_text"]
    return generated_text

iface = gr.Interface(
    fn=generate,
    inputs=["text", "text", gr.inputs.Slider(0.1, 2.0), gr.inputs.Slider(100, 50000), gr.inputs.Slider(0.1, 1.0)],
    outputs="text",
    title="Text Generation"
)
iface.launch()