← all papers · overview

Speculating Experts Accelerates Inference For Mixture-of-experts

Abstract

Mixture-of-Experts (MoE) models have gained popularity as a means of scaling the capacity of large language models (LLMs) while maintaining sparse activations and reduced per-token compute. However, in memory-constrained inference settings, expert weights must be offloaded to CPU, creating a performance bottleneck from CPU-GPU transfers during decoding. We propose an expert prefetching scheme that

Related papers

Ranked by semantic similarity — how closely each paper's abstract matches this one (100% = near-identical topic).