app.py

from huggingface_hub import InferenceClient
import gradio as gr

client = InferenceClient(
    model="https://qynvq9pllv2plc0v.us-east-1.aws.endpoints.huggingface.cloud"
)


def format_prompt(message, history):
  prompt = ""
  for user_prompt, bot_response in history:
    prompt += f"GPT4 Correct User: {user_prompt}<|end_of_turn|>GPT4 Correct Assistant: {bot_response}<|end_of_turn|>"
  prompt += f"GPT4 Correct User: {message}<|end_of_turn|>GPT4 Correct Assistant:"
  return prompt

def generate(
    prompt, history, temperature=0.9, max_new_tokens=1024, 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(f"{prompt}", history)
    stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
    output = ""

    for response in stream:
        if response.token.text=="<|end_of_turn|>":
            break
        output += response.token.text
        yield output
    return output


additional_inputs=[
    gr.Slider(
        label="Temperature",
        value=0.1,
        minimum=0.0,
        maximum=1.0,
        step=0.05,
        interactive=True,
        info="Higher values produce more diverse outputs",
    ),
    gr.Slider(
        label="Max new tokens",
        value=1024,
        minimum=0,
        maximum=1048,
        step=64,
        interactive=True,
        info="The maximum numbers of new tokens",
    ),
    gr.Slider(
        label="Top-p (nucleus sampling)",
        value=0.90,
        minimum=0.0,
        maximum=1,
        step=0.05,
        interactive=True,
        info="Higher values sample more low-probability tokens",
    ),
    gr.Slider(
        label="Repetition penalty",
        value=1.2,
        minimum=1.0,
        maximum=2.0,
        step=0.05,
        interactive=True,
        info="Penalize repeated tokens",
    )
]

examples=[["what is self realization according to bhagwan ramana maharishi", None, None, None, None, None, ],
          ["How does the teaching of bhagwan ramana maharishi hold good in the bay area for an aspiring startup founder", None, None, None, None, None,],
          ["How to teach a 8 year old about ramana maharishi's teaching", None, None, None, None, None,],
          ["why don't have the realization of the self like ramana maharishi as a default feature in us , is it not very inefficient for us to realize over the adulthood?", None, None, None, None, None,],
         ]

gr.ChatInterface(
    fn=generate,
    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel"),
    additional_inputs=additional_inputs,
    title="Starling Beta (Research Preview can make mistakes)",
    examples=examples,
    concurrency_limit=50,
).launch(show_api=False)