GitHub

AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

AWQ search for accurate quantization. Pre-computed AWQ model zoo for LLMs (LLaMA, Llama2, OPT, CodeLlama, StarCoder, Vicuna, LLaVA; load to generate quantized weights). Memory-efficient 4-bit Linear ...
GitHub

motokimura/timm_quantization_fx

Sensitivity analysis (and partial quantization) example is also provided. The figure below shows per-layer sensitivity analysis result of efficientnet_lite0 model. Only the static post-training ...
IEEE

Analysis of Quantization Noise in Fixed-Point HDFT Algorithms

Abstract: The Discrete Fourier Transform (DFT) algorithm is widely used in signal processing and communication systems to transform the signal to the frequency-domain. As real-time signal analysis is ...
IEEE

A Quantization Approach for the Reduced Size of Large Language Models

Abstract: The use of large-language models is widespread in a range of applications, including natural language processing and multimodal tasks. However, these models are computationally intensive.