paper-with-me

Papers

TeLLMe v2: An Efficient End-to-End Ternary LLM Prefill and Decode Accelerator with Table-Lookup Matmul on Edge FPGAs

2025-10-03 · Ye Qiao, Zhiheng Chen, Yifan Zhang, Yian Wang, Sitao Huang arxiv

With the emergence of wearable devices and other embedded systems, deploying large language models (LLMs) on edge platforms has become an urgent need. However, this is challenging because of their high computational and memory demands. Although recent low-bit quantization methods (e.g., BitNet, DeepSeek) compress weights to as low as 1.58~bits with minimal accuracy loss, edge deployment is still constrained by limited on-chip resources, power budgets, and the often-neglected long latency of the prefill stage. We present \textbf{TeLLMe}, the first table-lookup-based ternary LLM accelerator for low-power edge FPGAs that fully supports both prefill and autoregressive decoding using 1.58-bit weights and 8-bit activations. TeLLMe incorporates several novel techniques, including (1) a table-lookup-based ternary matrix multiplication (TLMM) engine utilizing grouped activations and online precomputation for low resource utilization and high throughput; (2) a fine-grained analytic URAM-based weight buffer management scheme for efficient loading and compute engine access; (3) a streaming dataflow architecture that fuses floating-point element-wise operations with linear computations to hide latency; (4) a reversed-reordered prefill stage attention with fused attention operations for high memory efficiency; and (5) a resource-efficient specialized decoding stage attention. Under a 5~W power budget, TeLLMe delivers up to 25~tokens/s decoding throughput and 0.45--0.96~s time-to-first-token (TTFT) for 64--128 token prompts, marking a significant energy-efficiency advancement in LLM inference on edge FPGAs.

📄 PDF Abstract BibTeX arXiv:2510.15926

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

TeLLMe: An Energy-Efficient Ternary LLM Accelerator for Prefilling and Decoding on Edge FPGAs

2025-04-22 · Ye Qiao, Zhiheng Chen, Yifan Zhang, Yian Wang 외

Deploying large language models (LLMs) on edge platforms is challenged by their high computational and memory demands. Although recent low-bit quantization methods (e.g., BitNet, DeepSeek) compress weights to as little a…

Quantization

Prefill/Decode-Aware Evaluation of LLM Inference on Emerging AI Accelerators

2026-06-14 · Shun Usami, Venkatram Vishwanath, E. Wes Bethel arxiv

As large language models (LLMs) are increasingly deployed in latency- and cost-sensitive settings, inference efficiency has become a central systems challenge. While GPUs dominate current deployments, a growing number of…

HALO: Memory-Centric Heterogeneous Accelerator with 2.5D Integration for Low-Batch LLM Inference

2025-10-03 · Shubham Negi, Kaushik Roy arxiv

The rapid adoption of Large Language Models (LLMs) has driven a growing demand for efficient inference, particularly in latency-sensitive applications such as chatbots and personalized assistants. Unlike traditional deep…

FlexLLM: Composable HLS Library for Flexible Hybrid LLM Accelerator Design

2026-01-22 · Jiahao Zhang, Zifan He, Nicholas Fraser, Michaela Blott 외 arxiv

We present FlexLLM, a composable High-Level Synthesis (HLS) library for rapid development of domain-specific LLM accelerators. FlexLLM exposes key architectural degrees of freedom for stage-customized inference, enabling…

TiM-DNN: Ternary in-Memory accelerator for Deep Neural Networks

2019-09-15 · Shubham Jain, Sumeet Kumar Gupta, Anand Raghunathan

The use of lower precision has emerged as a popular technique to optimize the compute and storage requirements of complex Deep Neural Networks (DNNs). In the quest for lower precision, recent studies have shown that tern…

GPUImage ClassificationLanguage Modelling