DSPE: An Energy-Efficient Edge Processor for DeepSeek Inference with MerkleTree-based Incremental Pruning, Multi-Stage Boothing Lookup and Dynamic Adaptive Posit Processing
In recent years, DeepSeek has achieved strong inference performance but remains hard to deploy on energy-constrained edge devices. This paper presents the DeepSeek Processing Element (DSPE), an edge-oriented architecture that alleviates the model's heavy computational and energy demands. DSPE introduces three techniques: the MerkleTree-based Incremental Pruning Scheme (MIPS) for secure redundant-vector reduction, the Multi-Stage Boothing Lookup Method (MBLM) for bit-flip-aware approximate multiplication, and the Dynamic Adaptive Posit Processing Mechanism (DAPPM), which introduces a new DA-Posit format and its corresponding hardware multiplication architecture. Implemented in TSMC 28nm CMOS, DSPE achieves 109.4 TFLOPS/W energy efficiency compared with state-of-the-art designs and offers a scalable foundation for edge deployment.
Cite
@article{arxiv.2605.08615,
title = {DSPE: An Energy-Efficient Edge Processor for DeepSeek Inference with MerkleTree-based Incremental Pruning, Multi-Stage Boothing Lookup and Dynamic Adaptive Posit Processing},
author = {Yuhan Zhang and Zhou Wang and Zhou Shu and Jiuren Zhou and Yanqing Xu and Xiaonan Tang and Shushan Qiao and Tianchun Ye and Yang Liu and Anil A. Bharath and Emm Mic Drakakis},
journal= {arXiv preprint arXiv:2605.08615},
year = {2026}
}
Comments
Accepted by DAC 2026, Long Beach, CA, USA. 7 pages, 8 figures. Corresponding author: Zhou Wang. {\dag}These authors contributed equally to this work: Yuhan Zhang and Zhou Wang