The paper investigates whether training Mixture-of-Experts (MoE) routers can improve memory‑bandwidth locality on consumer GPUs. Using a new zero‑surgery telemetry tool, the authors measure that a large Qwen3‑235B model is bottlenecked by disk‑based expert access, and that an LRU cache can serve a majority of requests. They pre‑register experiments training 137 M‑parameter MoE models with locality‑aware losses, finding that while cache misses can drop up to 60 % (99 % static‑pin hit rate), every configuration fails to meet a strict 1 % perplexity threshold, indicating a tight coupling between cache efficiency and model quality.
By Shriniwas Ramesh Suram
Serving a 235B-parameter Mixture-of-Experts (MoE) model on a single 8 GB GPU is bottlenecked not by compute but by memory bandwidth: decode must stream each token's active experts from whichever tier holds them, and on consumer hardware most experts sit on an SSD far slower than RAM. We quantify this bandwidth wall on Qwen3-235B (Q4_K_M, 134 GB): measured decode is 0.
The paper introduces a cache‑aware post‑training framework for Mixture‑of‑Experts (MoE) models that jointly adapts the MoE backbone and lightweight auxiliary cache routers while keeping the native Top‑K expert‑selection rule. Two modes are proposed: Temporal Router, which predicts same‑layer reuse and retains experts for future tokens, and Spatio‑Temporal Router, which adds a Spatio Router that refines the temporal cache using the causal predecessor’s hidden state. Experiments on Qwen3 and GPT‑OSS across GSM8K, MATH, and CommonsenseQA show that Temporal Router improves cache hit rates and reduces expert‑weight traffic, while Spatio‑Temporal Router achieves the best load‑adjusted efficiency, outperforming strong prefetching baselines.
By Zhenhe Wu, Yaping Jin, Qinghua Xing, Hang Zhou, Wei He, Xianjie Wu, Xianfu Cheng, Jian Yang, Hanting Chen
arXiv:2607. 24787v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) models expand foundation model capacity through conditional expert activation, but their full expert pools remain difficult to deploy under limited accelerator memory.
By Jinwei Kong, Runqi Meng, Fanyi Wang, Wentao Qiu, Haotian Hu, Yongjian Zhou, Zhenhua Ge
arXiv:2607. 24434v1 Announce Type: cross Abstract: Large Mixture-of-Experts (MoE) language models are attractive for end-device deployment because only a small subset of experts is active per token, but their routed expert weights often exceed accelerator memory.
By Dengke Han
MaskCoFT introduces a masked co‑adaptive fine‑tuning approach for mixture‑of‑experts language models, training both routers and experts jointly with a cross‑entropy loss. A learnable binary mask limits each layer’s Top‑K routing to a subset of experts during fine‑tuning, allowing experts to adapt to the tokens they receive. In simulated GPU cache scenarios, MaskCoFT reduces expert fetches per token by 23.7% for Mixtral‑8x7B and 10.1% for DeepSeek‑V2‑Lite, and lowers inference time per output token by up to 16.4% and 5.5% respectively, while maintaining or improving accuracy across nine benchmarks.
By Junfeng Wu, Zehao Fan, Hadjer Benmeziane, Kaoutar El Maghraoui, Liu Liu, Yinan Wang
WiSP (Working‑Set Paging) is a routing‑aware expert pager that allows Mixture‑of‑Experts models to run on GPUs that cannot hold the entire expert pool by paging experts in and out of VRAM while preserving byte‑identical outputs. On a 24 GiB RTX 3090, WiSP doubles decode throughput compared to static offload when the model does not fit, and its companion policy MV‑WSA allocates VRAM between resident experts and KV cache based on marginal latency benefit, reducing end‑to‑end time by up to 1.19× without altering model outputs.
By Jiamu Zhang, Liang Wu, Mayank Darbari, Liangjie Hong
The paper investigates the performance impact of using a remote KV cache tier for long‑context language model serving. In a disaggregated deployment of GLM‑5.3 on 8xB300 nodes, enabling the remote tier initially increased the 16k‑token p99 latency from 3.96 s to 17.3 s, and synchronous loading caused severe queue instability. By sharding the KV store across context‑parallel ranks, partitioning the remote tier, adding asynchronous loading, and fixing client bugs, the authors reduced the p99 latency from 83.32 s to 0.67 s and increased prefill throughput by 14–34 %.
whyItMatters":"The study demonstrates that careful sharding and asynchronous handling of remote KV caches can dramatically improve latency and throughput in large‑scale LLM serving systems."
By Muhammad Ibrahim
Dynamic Expert Quantization (DynaExq) is a runtime-aware mixed-precision serving system designed for single‑GPU Mixture‑of‑Experts (MoE) inference under a hard high‑bandwidth memory (HBM) envelope. It treats the problem as an online, budget‑constrained precision allocation task, keeping the most frequently used experts at higher precision while relegating the rest to low‑precision fallbacks. By estimating expert hotness from router traces and asynchronously promoting or demoting experts, DynaExq maintains a fully materialized expert set during the forward pass, improving accuracy and throughput compared to static post‑training quantization and offloading/prefetch baselines.
whyItMatters":"DynaExq enables efficient deployment of large MoE models on memory‑limited GPUs by dynamically allocating precision based on runtime expert usage, thereby reducing memory footprint and latency while boosting accuracy and throughput."
By Kexin Chu, Dawei Xiang, Zixu Shen, Yiwei Yang, Zecheng Liu, Wei Zhang
arXiv:2608. 02989v1 Announce Type: new Abstract: Speculative decoding verifies a tree of draft tokens in one target-model forward pass.
By Shuang Liang (Mark), Hao (Mark), Chen, Zhiwen Mo, Qianzhou Wang, Guoyu Li, Lingxiao Ma, Wayne Luk
arXiv:2609.36222v1 Announce Type: new
Abstract: Large language models are increasingly expensive to serve. In large-scale serving systems, autoregressive decoding is often bottlenecked by transferrin...
By Ali Abbasi, Justin Shi, Soheil Kolouri
arXiv:2608. 15383v1 Announce Type: new Abstract: Sparse mixture-of-experts (MoE) language models reduce arithmetic by activating only a small subset of experts per token, yet deployment still requires storing and moving the full expert bank.
By Amjad Saab