FlexQuant: Elastic Quantization Framework for Locally Hosted LLM on Edge Devices
This work addresses memory elasticity for locally hosted LLMs on edge devices, offering a domain-specific solution with incremental improvements in quantization and pruning.
The paper tackles the challenge of deploying large language models on edge devices with fluctuating memory by introducing FlexQuant, an elastic quantization framework that achieves a 15x improvement in transition granularity and a 10x reduction in storage compared to state-of-the-art methods.
Deploying LLMs on edge devices presents serious technical challenges. Memory elasticity is crucial for edge devices with unified memory, where memory is shared and fluctuates dynamically. Existing solutions suffer from either poor transition granularity or high storage costs. We propose FlexQuant, a novel elasticity framework that generates an ensemble of quantized models, providing an elastic hosting solution with 15x granularity improvement and 10x storage reduction compared to SoTA methods. FlexQuant works with most quantization methods and creates a family of trade-off options under various storage limits through our pruning method. It brings great performance and flexibility to the edge deployment of LLMs.