xmadai/Mistral-Large-Instruct-2407-xMADai-INT4

This model has been xMADified!

This repository contains mistralai/Mistral-Large-Instruct-2407 quantized from 16-bit floats to 4-bit integers, using xMAD.ai proprietary technology.

Why should I use this model?

1. Memory-efficiency: The full-precision model is around 250 GB, while this xMADified model is only 65 GB, making it feasible to run on a single 80 GB GPU or 2x 40 GB GPUs.

2. Accuracy: This xMADified model preserves the quality of the full-precision model. In the table below, we present the zero-shot accuracy on popular benchmarks of this xMADified model against the GPTQ-quantized model. The xMADai model offers higher accuracy than the GPTQ model.

Model	MMLU STEM	MMLU Humanities	MMLU Social Sciences	MMLU Other	LAMBADA Standard	LAMBADA OpenAI
GPTQ Mistral-Large-Instruct-2407	77.26	77.83	89.57	86.03	74.95	81.04
xMADai Mistral-Large-Instruct-2407 (this model)	77.26	77.98	89.57	86.26	75.20	81.29

3. Fine-tuning: These models are fine-tunable over reduced hardware in mere 3-clicks. Watch our product demo here

How to Run Model

Loading the model checkpoint of this xMADified model requires 65 GB of VRAM. Hence it can be efficiently run on 2x 40 GB GPUs.

Package prerequisites: Run the following commands to install the required packages.

pip install torch==2.4.0  # Run following if you have CUDA version 11.8: pip install torch==2.4.0 --index-url https://download.pytorch.org/whl/cu118
pip install transformers accelerate optimum
pip install -vvv --no-build-isolation "git+https://github.com/PanQiWei/[email protected]"

Sample Inference Code

from transformers import AutoTokenizer
from auto_gptq import AutoGPTQForCausalLM

model_id = "xmadai/Mistral-Large-Instruct-2407-xMADai-INT4"
prompt = [
    {"role": "system", "content": "You are a helpful assistant, that responds as a pirate."},
    {"role": "user", "content": "What's Deep Learning?"},
]

tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=False)

inputs = tokenizer.apply_chat_template(
    prompt,
    tokenize=True,
    add_generation_prompt=True,
    return_tensors="pt",
    return_dict=True,
).to("cuda")

model = AutoGPTQForCausalLM.from_quantized(
    model_id,
    device_map='auto',
    trust_remote_code=True,
)

outputs = model.generate(**inputs, do_sample=True, max_new_tokens=1024)
print(tokenizer.batch_decode(outputs, skip_special_tokens=True))

Citation

If you found this model useful, please cite our research paper.

@article{zhang2024leanquant,
  title={LeanQuant: Accurate and Scalable Large Language Model Quantization with Loss-error-aware Grid},
  author={Zhang, Tianyi and Shrivastava, Anshumali},
  journal={arXiv preprint arXiv:2407.10032},
  year={2024},
  url={https://arxiv.org/abs/2407.10032},
}

Contact Us

For additional xMADified models, access to fine-tuning, and general questions, please contact us at [email protected] and join our waiting list.