Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2501.07139 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915100828368896 |
|---|---|
| author | Chai, Yuji Kwen, Mujin Brooks, David Wei, Gu-Yeon |
| author_facet | Chai, Yuji Kwen, Mujin Brooks, David Wei, Gu-Yeon |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_07139 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FlexQuant: Elastic Quantization Framework for Locally Hosted LLM on Edge Devices Chai, Yuji Kwen, Mujin Brooks, David Wei, Gu-Yeon Artificial Intelligence Performance 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. |
| title | FlexQuant: Elastic Quantization Framework for Locally Hosted LLM on Edge Devices |
| topic | Artificial Intelligence Performance |
| url | https://arxiv.org/abs/2501.07139 |