import gradio as gr
from transformers import pipeline

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
model_name = "Joetib/en-twi-llama-3.2-1B"

from transformers import pipeline

p = pipeline("text-generation", model=model_name, device_map="auto")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]
    message = f"<s> {system_message} [INST] {message} [/INST]"

    

    final = ""

    for response   in  p(message,
        max_new_tokens=20,
        do_sample=True,
        temperature=1,
    ):
        text = response["generated_text"].split("[/INST]")[-1]
        final += text

    return final

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="Translate to Twi", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)


if __name__ == "__main__":
    demo.launch()