Large Language Models (LLMs) from the GPT family have become extremely popular, leading to a race towards reducing their inference costs to allow for efficient local computation. However, the vast majority of existing work focuses on weight-only quantization, which can reduce runtime costs in the memory-bound one-token-at-a-time generative setting, but does not address costs in compute-bound scenarios, such as batched inference or prompt processing.In this paper, we address the general quantization problem, where both weights and activations should be quantized, which leads to computational improvements in general. We show that the majority of inference computations for large generative models can be performed with both weights and activations being cast to 4 bits, while at the same time maintaining good accuracy. We achieve this via a hybrid quantization strategy called QUIK that compresses most of the weights and activations to 4-bit, while keeping a small fraction of “outlier” weights and activations in higher-precision. QUIK is that it is designed with computational efficiency in mind: we provide GPU kernels matching the QUIK format with highly-efficient layer-wise runtimes, which lead to practical end-to-end throughput improvements of up to 3.4x relative to FP16 execution. We provide detailed studies for models from the OPT, LLaMA-2 and Falcon families, as well as a first instance of accurate inference using quantization plus 2:4 sparsity.Anonymized code is available.

In this paper, we consider the problem of sparsifying BERT models, which are a key building block for natural language processing, in order to reduce their storage and computational cost. We introduce the Optimal BERT Surgeon (oBERT), an efficient and accurate pruning method based on approximate second-order information, which we show to yield state-of-the-art results in both stages of language tasks: pre-training and fine-tuning. Specifically, oBERT extends existing work on second-order pruning by allowing for pruning weight blocks, and is the first such method that is applicable at BERT scale. Second, we investigate compounding compression approaches to obtain highly compressed but accurate models for deployment on edge devices. These models significantly push boundaries of the current state-of-the-art sparse BERT models with respect to all metrics: model size, inference speed and task accuracy. For example, relative to the dense BERT-base, we obtain 10x model size compression with < 1% accuracy drop, 10x CPU-inference speedup with < 2% accuracy drop, and 29x CPU-inference speedup with < 7.5% accuracy drop. Our code, fully integrated with Transformers and SparseML, is available at https://github.com/neuralmagic/sparseml/tree/main/research/optimal_BERT_surgeon_oBERT.