app.py

import gradio as gr
from huggingface_hub import InferenceClient
import spaces

import os
import torch
import transformers
from transformers import (
  AutoTokenizer,
  AutoModelForCausalLM,
  BitsAndBytesConfig,
  pipeline
)

#################################################################
# Tokenizer
#################################################################

model_name='kimou605/shadow-clown-BioMistral-7B-DARE'
model_config = transformers.AutoConfig.from_pretrained(
    model_name,
)

tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
tokenizer.pad_token = tokenizer.eos_token
tokenizer.padding_side = "right"

#################################################################
# bitsandbytes parameters
#################################################################

# Activate 4-bit precision base model loading
use_4bit = True

# Compute dtype for 4-bit base models
bnb_4bit_compute_dtype = "float16"

# Quantization type (fp4 or nf4)
bnb_4bit_quant_type = "nf4"

# Activate nested quantization for 4-bit base models (double quantization)
use_nested_quant = True

#################################################################
# Set up quantization config
#################################################################
compute_dtype = getattr(torch, bnb_4bit_compute_dtype)

bnb_config = BitsAndBytesConfig(
    load_in_4bit=use_4bit,
    bnb_4bit_quant_type=bnb_4bit_quant_type,
    bnb_4bit_compute_dtype=compute_dtype,
    bnb_4bit_use_double_quant=use_nested_quant,
)

# Check GPU compatibility with bfloat16
if compute_dtype == torch.float16 and use_4bit:
    major, _ = torch.cuda.get_device_capability()
    if major >= 8:
        print("=" * 80)
        print("Your GPU supports bfloat16: accelerate training with bf16=True")
        print("=" * 80)

#################################################################
# Load pre-trained config
#################################################################
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    quantization_config=bnb_config,
)

pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
    tokenizer=tokenizer,

)

# messages = [{"role": "user", "content": "what is ai"}]
# prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
# outputs = pipeline(prompt, max_new_tokens=1024, do_sample=True, temperature=0.01, top_k=1, top_p=0.001)[0]["generated_text"]
# print(outputs[0]["generated_text"])


@spaces.GPU
def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    subprocess.check_call([sys.executable, "-m", "pip", "install", "-U", "bitsandbytes"])
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    response = ""

    for message in pipeline.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message.choices[0].delta.content

        response += token
        yield response


demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", 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)",
        ),
    ],
    css="footer{display:none !important}",
)


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