🛰️ Daily AI Frontier
‹ back to 2026-07-17

ExaGEMM: Exploration Framework for CPU-Driven ML Inference via Associative In-Register Computing for Low-Bit GEMM

Research Efficiency & Systems

Merged summary

TL;DR - ExaGEMM is a codesign and exploration framework for running low-bit GEMM efficiently on CPUs via register-resident lookup-table (LUT) execution, addressing the poor fit between very-low-bit ML inference and conventional CPU datapaths. It matters because low-bit and mixed-precision GEMM is increasingly central to efficient LLM inference, yet CPU support has been ad hoc.

  • Targets fragmented low-bit regimes (1/2/4-bit weights with varying activation precision) under fixed SIMD and register-file budgets, framing lightweight CPU support selection as a first-class design problem.
  • Key insight: existing SIMD datapaths already handle table generation and accumulation, so the only new hardware needed is an in-register select/feed mechanism with explicitly modeled cost.
  • Uses analytical models (register feasibility, compute cost, memory traffic, hardware overhead) to prune the candidate space by 99.2% before simulation, then emits ISA specs, gem5 patches, and GEMM kernels for validation.
  • Reports 13.29x latency improvement over software-only baselines, with workload-aware frontier selection especially beneficial for mixed-precision LLM workloads.

Sources (1)

ExaGEMM: Exploration Framework for CPU-Driven ML Inference via Associative In-Register Computing for Low-Bit GEMM

arXiv cs.AR Hyunwoo Oh, Suyeon Jang, Hanning Chen, Sanggeon Yun, Ryozo Masukawa, Mohsen Imani 2026-07-16 arXiv:2607.14622

TL;DR - ExaGEMM is a codesign and exploration framework for running low-bit GEMM efficiently on CPUs via register-resident lookup-table (LUT) execution, addressing the poor fit between very-low-bit ML inference and conventional CPU datapaths. It matters because low-bit and mixed-precision GEMM is increasingly central to efficient LLM inference, yet CPU support has been ad hoc.

  • Targets fragmented low-bit regimes (1/2/4-bit weights with varying activation precision) under fixed SIMD and register-file budgets, framing lightweight CPU support selection as a first-class design problem.
  • Key insight: existing SIMD datapaths already handle table generation and accumulation, so the only new hardware needed is an in-register select/feed mechanism with explicitly modeled cost.
  • Uses analytical models (register feasibility, compute cost, memory traffic, hardware overhead) to prune the candidate space by 99.2% before simulation, then emits ISA specs, gem5 patches, and GEMM kernels for validation.
  • Reports 13.29x latency improvement over software-only baselines, with workload-aware frontier selection especially beneficial for mixed-precision LLM workloads.
item →