from huggingface_hub import InferenceClient
import gradio as gr

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

# Your system prompt
SYSTEM_PROMPT = "Rewrite this prompt in sentence form without changing the essence"

def format_prompt(message, history=[]):
    prompt = "<s>"
    prompt += f"[INST] SYSTEM: {SYSTEM_PROMPT} [/INST]"  # Add the system prompt here
    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, 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(prompt)

    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.strip('</s>')
    return output.strip('</s>')

with gr.Blocks() as demo:
    input_text = gr.Textbox(placeholder="Enter your prompt here...", lines=2, max_lines=2, label="Prompt")
    submit_button = gr.Button("Generate")
    output_text = gr.Textbox(label="Output", interactive=True, lines=10)
    
    
    submit_button.click(fn=generate, inputs=input_text, outputs=output_text)

demo.launch()