MobileMoE: Scaling On-Device Mixture of Experts

• URL: http://arxiv.org/abs/2605.27358v1
• Date: Tue, 26 May 2026 17:58:24 GMT
• Title: MobileMoE: Scaling On-Device Mixture of Experts
• Abstract summary: We present MobileMoE, a family of on-device MoE language models with sub-billion active parameters.<n>We first formulate an on-device MoE scaling law that jointly optimize MoE architecture under mobile memory and compute constraints.<n>We train MobileMoE with a four-stage recipe covering pre-training, mid-training, instruction fine-tuning, and quantization-aware training.
• Score: 32.364288082407406
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Mixture-of-Experts (MoE) has become the de facto architecture for hundred-billion-parameter language models, yet its advantages at sub-billion scales for on-device deployment remain largely unexplored. To close this gap, we present MobileMoE, a family of on-device MoE language models with sub-billion active parameters (0.3-0.9B active and 1.3-5.3B total) that establish a new Pareto frontier for on-device LLMs. We first formulate an on-device MoE scaling law that jointly optimizes MoE architecture under mobile memory and compute constraints, identifying an on-device sweet spot - moderate sparsity with fine-grained and shared experts - that is simultaneously memory and compute-optimal. Building on the derived architectures, we train MobileMoE with a four-stage recipe covering pre-training, mid-training, instruction fine-tuning, and quantization-aware training, all on open-source datasets. Across 14 benchmarks, MobileMoE matches or exceeds leading on-device dense LLMs with 2-4$\times$ fewer inference FLOPs, and matches or surpasses the state-of-the-art MoE OLMoE-1B-7B with up to 60% fewer parameters. To bridge the last mile to mobile deployment, we provide the first efficient MoE inference on commodity smartphones with comprehensive on-device profiling. At comparable INT4 weight memory, MobileMoE-S delivers $1.8$-$3.8\times$ faster prefill and $2.2$-$3.4\times$ faster decode than the dense baseline MobileLLM-Pro.

Related papers

• MeMo: Memory as a Model [96.54179792557863]
  Large language models (LLMs) achieve strong performance across a wide range of tasks, but remain frozen after pretraining until subsequent updates.<n>MeMo is a modular framework that encodes new knowledge into a dedicated memory model while keeping the LLM parameters unchanged.<n>MeMo achieves strong performance compared to existing methods across diverse settings.
  arXiv  Detail & Related papers  (2026-05-14T17:51:34Z)
• Mobile-O: Unified Multimodal Understanding and Generation on Mobile Device [90.46496321553843]
  We present Mobile-O, a compact vision-language-diffusion model that brings unified multimodal intelligence to a mobile device.<n>Its core module, the Mobile Conditioning Projector (MCP), fuses vision-language features with a diffusion generator using depthwise-separable convolutions and layerwise alignment.<n>Running in only 3s per 512x512 image on an iPhone, Mobile-O establishes the first practical framework for real-time unified multimodal understanding and generation on edge devices.
  arXiv  Detail & Related papers  (2026-02-23T18:59:58Z)
• FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge Devices [0.0]
  Mixture-of-Experts (MoE) models have gained attention for efficiently scaling large language models.<n>MoE models are extremely large, their sparse activation enables inference to be performed by accessing only a fraction of the model at a time.<n>We propose FlashMoE, a system that offloads inactive experts to SSD, enabling efficient MoE inference under limited RAM.
  arXiv  Detail & Related papers  (2026-01-22T17:07:33Z)
• SonicMoE: Accelerating MoE with IO and Tile-aware Optimizations [54.303301888915406]
  Mixture of Experts (MoE) models have emerged as the de facto architecture for scaling up language models without significantly increasing the computational cost.<n>We propose a memory-efficient algorithm to compute the forward and backward passes of MoEs with minimal activation caching.<n>We also propose a novel "token rounding" method that minimizes the wasted compute due to padding in Grouped GEMM kernels.
  arXiv  Detail & Related papers  (2025-12-16T04:39:10Z)
• SlimMoE: Structured Compression of Large MoE Models via Expert Slimming and Distillation [82.53411922988039]
  We introduce SlimMoE, a multi-stage compression framework for transforming large MoE models into much smaller, efficient variants.<n>Using this framework, we compress Phi 3.5-MoE (41.9B total/6.6B activated parameters) to create Phi-mini-MoE (7.6B total/2.4B activated parameters) and Phi-tiny-MoE (3.8B total/1.1B activated parameters)<n>Our experiments demonstrate that these compressed models outperform others of similar size and remain competitive with larger models.
  arXiv  Detail & Related papers  (2025-06-23T07:15:59Z)
• FSMoE: A Flexible and Scalable Training System for Sparse Mixture-of-Experts Models [21.96960353910023]
  We introduce FSMoE, a flexible training system optimizing task scheduling with three novel techniques.<n>We conduct extensive experiments with configured MoE layers and real-world MoE models on two GPU clusters.<n> FSMoE supports four popular types of MoE routing functions and is more efficient than existing implementations.
  arXiv  Detail & Related papers  (2025-01-18T10:14:37Z)
• Dense Training, Sparse Inference: Rethinking Training of Mixture-of-Experts Language Models [62.4691912312317]
  Mixture-of-Experts (MoE) language models can reduce computational costs by 2-4$times$ compared to dense models without sacrificing performance. We propose a hybrid dense training and sparse inference framework for MoE models (DS-MoE) which achieves strong computation and parameter efficiency.
  arXiv  Detail & Related papers  (2024-04-08T14:39:49Z)
• MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases [46.997172696192195]
  This paper addresses the need for efficient large language models (LLMs) on mobile devices, driven by increasing cloud costs and latency concerns. We focus on designing top-quality LLMs with fewer than a billion parameters, a practical choice for mobile deployment.
  arXiv  Detail & Related papers  (2024-02-22T18:58:55Z)
• Mobile V-MoEs: Scaling Down Vision Transformers via Sparse Mixture-of-Experts [55.282613372420805]
  We explore the use of sparse MoEs to scale-down Vision Transformers (ViTs) to make them more attractive for resource-constrained vision applications. We propose a simplified and mobile-friendly MoE design where entire images rather than individual patches are routed to the experts. We empirically show that our sparse Mobile Vision MoEs (V-MoEs) can achieve a better trade-off between performance and efficiency than the corresponding dense ViTs.
  arXiv  Detail & Related papers  (2023-09-08T14:24:10Z)
• MoESys: A Distributed and Efficient Mixture-of-Experts Training and Inference System for Internet Services [32.278096820269816]
  We present a novel MoESys that boosts efficiency in both large-scale training and inference. Specifically, in the training procedure, the proposed MoESys adopts an Elastic MoE training strategy with 2D prefetch and Fusion communication over Hierarchical storage. For scalable inference in a single node, MoESys builds the CPU-GPU memory jointly into a ring of sections to load the model, and executes the computation tasks across the memory sections in a round-robin manner for efficient inference.
  arXiv  Detail & Related papers  (2022-05-20T09:09:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.