app.py

import os
import gradio as gr
from huggingface_hub import InferenceClient

HF_TOKEN = os.environ.get("HF_TOKEN", None)
API_URL = "meta-llama/Llama-2-70b-chat-hf"
API_URL_2 = "codellama/CodeLlama-34b-Instruct-hf"
BOT_NAME = "Assistant"

STOP_SEQUENCES = ["\nUser:", " User:", "###", "</s>"]

EXAMPLES = [
    ["Hey LLAMA! Any recommendations for my holidays in Abu Dhabi?"],
    ["What's the Everett interpretation of quantum mechanics?"],
    ["Give me a list of the top 10 dive sites you would recommend around the world."],
    ["Can you tell me more about deep-water soloing?"],
    ["Can you write a short tweet about the release of our latest AI model, LLAMA LLM?"]
    ]

client = InferenceClient(
    API_URL,
    token=HF_TOKEN,
)

client2 = InferenceClient(
    API_URL_2,
    token=HF_TOKEN,
)

def format_prompt(message, history, system_prompt):
  prompt = ""
  if system_prompt:
    prompt += f"System: {system_prompt}\n"
  for user_prompt, bot_response in history:
    prompt += f"User: {user_prompt}\n"
    prompt += f"{BOT_NAME}: {bot_response}\n"
  prompt += f"""User: {message}\n{BOT_NAME}:"""
  return prompt

seed = 42

def generate(
    prompt, history, system_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)
    global seed
    generate_kwargs = dict(
        temperature=temperature,
        max_new_tokens=max_new_tokens,
        top_p=top_p,
        repetition_penalty=repetition_penalty,
        stop_sequences=STOP_SEQUENCES,
        do_sample=True,
        seed=seed,
    )
    seed = seed + 1
    formatted_prompt = format_prompt(prompt, history, system_prompt)

    cli = client
    status = cli.get_model_status()
    print(f"Model 1 status: {status}")
    if not status.loaded and status.state == 'Loadable':
        cli.text_generation('Hello!', **generate_kwargs, return_full_text=False)
        cli = client2

    try:
        stream = cli.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
        output = ""

        for response in stream:
            output += response.token.text
    
            for stop_str in STOP_SEQUENCES:
                if output.endswith(stop_str):
                    output = output[:-len(stop_str)]
#                    output = output.rstrip()
                    yield output
            yield output
    except Exception as e:
        raise gr.Error(f"Client error while generating: {e}")
    return output

additional_inputs=[
    gr.Textbox("", label="Optional system prompt"),
    gr.Slider(
        label="Temperature",
        value=0.9,
        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=256,
        minimum=0,
        maximum=3000,
        step=64,
        interactive=True,
        info="The maximum numbers of new tokens",
    ),
    gr.Slider(
        label="Top-p (nucleus sampling)",
        value=0.90,
        minimum=0.01,
        maximum=0.99,
        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",
    )
]

with gr.Blocks() as demo:
    
    gr.ChatInterface(
        generate, 
        examples=EXAMPLES,
        additional_inputs=additional_inputs,
    ) 

demo.queue(concurrency_count=100, api_open=False).launch(show_api=False)