app.py

from huggingface_hub import InferenceClient
import gradio as gr

client = InferenceClient(
    "mistralai/Mistral-7B-Instruct-v0.3"
)

# Your system prompt
SYSTEM_PROMPT = "Your goal is to create engaging, authentic, and contextually appropriate captions for social media platforms. The captions should captivate the audience without being cringe-worthy, ensuring they resonate well with diverse demographics."

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, history=[], 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, 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

iface = gr.Interface(
    fn=generate,
    inputs=[
        gr.Textbox(placeholder="Enter your prompt here...", lines=2, max_lines=2, label=""),
        gr.Button("Generate")
    ],
    outputs=gr.Textbox(label="Output", interactive=True, lines=10),
    layout="vertical"
)

iface.launch()