MobileMoE: Scaling On-Device Mixture of Experts
About
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 benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Commonsense Reasoning | WinoGrande | -- | 1442 | |
| Commonsense Reasoning | HellaSwag | HellaSwag Accuracy74.6 | 711 | |
| Mathematical Reasoning | MATH 500 | Top-1 Accuracy32.2 | 384 | |
| Commonsense Reasoning | PIQA | Accuracy80 | 213 | |
| Commonsense Reasoning | SIQA | Accuracy54.3 | 168 | |
| Knowledge | MMLU | Accuracy59.6 | 161 | |
| Scientific Reasoning | ARC Challenge | Accuracy57 | 115 | |
| Reasoning | GSM8K | -- | 111 | |
| Instruction Following | IFEval | -- | 89 | |
| Science Question Answering | OpenBookQA | Accuracy42.8 | 82 |