English

QIGen: Generating Efficient Kernels for Quantized Inference on Large Language Models

Machine Learning 2023-07-10 v1 Computation and Language Performance

Abstract

We present ongoing work on a new automatic code generation approach for supporting quantized generative inference on LLMs such as LLaMA or OPT on off-the-shelf CPUs. Our approach is informed by the target architecture and a performance model, including both hardware characteristics and method-specific accuracy constraints. Results on CPU-based inference for LLaMA models show that our approach can lead to high performance and high accuracy, comparing favorably to the best existing open-source solution. A preliminary implementation is available at https://github.com/IST-DASLab/QIGen.

Keywords

Cite

@article{arxiv.2307.03738,
  title  = {QIGen: Generating Efficient Kernels for Quantized Inference on Large Language Models},
  author = {Tommaso Pegolotti and Elias Frantar and Dan Alistarh and Markus Püschel},
  journal= {arXiv preprint arXiv:2307.03738},
  year   = {2023}
}
R2 v1 2026-06-28T11:24:45.919Z