cs.LGSep 8, 2026

Hyperparameter Scaling Laws Across MoE Sparsity

Authors: Changxin Tian, Kunlong Chen, Jia Liu, Ziqi Liu, Zhiqiang Zhang, Jun Zhou

Abstract

Mixture-of-Experts (MoE) models expand model capacity without a proportional increase in training compute, but increasing sparsity makes reliable hyperparameter transfer challenging. In this work, we show that conventional hyperparameter scaling laws are insufficient for ultra-sparse MoEs: the optimal learning rate and batch size vary with activation ratio, and these shifts cannot be explained by either total or activated parameter count alone. To characterize this dependence, we conduct 1,800 pre-training runs spanning six activated-parameter scales and models with up to 6B total non-embedding parameters, processing approximately 20 trillion tokens at a cost of 200,000 equivalent H800 GPU-hours. Our results reconcile conflicting findings in prior work by revealing two scaling regimes. At fixed sparsity, the optimal batch size follows a power-law relationship with training tokens DD, whereas the optimal learning rate scales with training compute CC and remains robust to the allocation between model size and data. Across sparsity levels, the activation ratio AA enters both relationships as an additional multiplicative power-law factor. These observations lead to unified hyperparameter scaling laws that transfer across MoE sparsity levels. Large-scale evaluation shows that the scaling form outperforms alternative functional forms. On a held-out ultra-sparse MoE with 12B total parameters and only 1/64 of its experts activated, the predicted hyperparameters remain close to the observed optima, supporting joint extrapolation across model scale and sparsity. Further experiments demonstrate transfer across expert granularities and isolate the effect of activation ratio from that of total expert count.

Explore similar work

CardsList
May 13, 2026cs.LG

How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization

Recent frontier large language models predominantly rely on Mixture-of-Experts (MoE) architectures. Despite empirical progress, there is still no principled understanding of how hyperparameters should scale with network width NN, expert width NeN_e, number of experts MM, sparsity KK, and depth LL to ensure both stability and optimal performance at scale. We take a principled step toward resolving this gap by analyzing three different scaling regimes: (I) co-scaling N≍NeN\asymp N_e, (II) co-scaling N≍M≍KN\asymp M\asymp K, and (III) full proportional scaling of N,Ne,MN, N_e, M, and KK. For each regime, we develop a novel Dynamical Mean Field Theory (DMFT) description of the limiting training dynamics of MoEs that provides a formal foundation for our analysis. Within this framework, we derive the unique parameterization for SGD and Adam satisfying all maximal-update (μμ) desiderata. We then show that the resulting μμP prescription does not reliably induce monotonic improvement with scale or robust learning-rate transfer. We trace these pathologies to scale-dependent observables in the aggregation dynamics, which motivates a refined set of desiderata that we term maximal scale stability. Guided by this principle, we derive a Maximally Scale-Stable Parameterization (MSSP) for both SGD and Adam in all three scaling regimes, and characterize the corresponding limiting dynamics - qualitatively distinct from the μμP limit - through a separate DMFT analysis. Experiments verify that MSSP robustly recovers learning rate transfer and monotonic improvement with scale across regimes. Combined with existing depth-scaling theory, these results provide a complete scaling prescription for MoE architectures as a function of width, depth, expert width, and number of experts.
Leena Chennuru Vankadara, Moritz Haas, Luke Hayward +2
Learning RatesScaling LawsLarge Language Model ScaleTransfer Learning
AskPDF
May 22, 2026cs.LG

Complete-muE: Optimal Hyperparameter Transfer and Scaling for MoE Models

We propose Complete-muE, a framework which targets hyperparameter transfer across dense FFN and any Mixture-of-Experts (MoE) setups in transformer blocks. Existing tools such as μμP (requires fixed architectue) or SDE (requires fixed per-step token count) cannot directly solve the hyperparameter transfer problem in MoE setups because Dense to MoE transfer or MoE total experts scaling changes both architecture and tokens per expert. Complete-muE solves this challenge with a two-bridge system: BridgeI maps between dense FFN and Dense MoE by active-width μμP with a normalized router scale. BridgeII maps between Dense MoE and sparse MoE by activated-expert scaling, where the first-order SDE LR/WD correction cancels while a bounded residual σ0σ_0 shift remains. The resulting transfer rule, which we term as Complete muE, covers changes in activated experts, total capacity, granularity, and shared/group-balanced hybrids for MoE models as well as network width/depth, batch size, and duration changes for general Transformer models. Extensive language model and diffusion model pretraining experiments confirm that complete-muE yields relatively stable hyperparameter optima across model architectures and parameter counts -- with only minor drift consistent with the non-strict SDE behavior of Bridge~II. In practice this drift is small enough that hyperparameters tuned on a single dense reference transfer near-optimally to all MoE configurations -- \emph{tune dense once, transfer to all} is the practical recipe at the core of Complete-muE. This enables MoE models to achieve accelerated convergence speedup over dense models when scaling model capacity without costly hyperparameter search.
Hongwu Peng, Ohiremen Dibua, Yuanjun Xiong +3
Transformer VariantParameter-Efficient Fine-Tuning MethodsHyperparameterLarge Language Model ScaleMuon
AskPDF
May 13, 2026cs.LG

EMO: Frustratingly Easy Progressive Training of Extendable MoE

Sparse Mixture-of-Experts (MoE) models offer a powerful way to scale model size without increasing compute, as per-token FLOPs depend only on k active experts rather than the total pool of E experts. Yet, this asymmetry creates an MoE efficiency paradox in practice: adding more experts balloons memory and communication costs, making actual training inefficient. We argue that this bottleneck arises in part because current MoE training allocates too many experts from the beginning, even though early-stage data may not fully utilize such capacity. Motivated by this, we propose EMO, a simple progressive training framework that treats MoE capacity as expandable memory and grows the expert pool over the course of training. EMO explicitly models sparsity in scaling law to derive stage-wise compute-optimal token budgets for progressive expansion. Empirical results show that EMO matches the performance of a fixed-expert setup in large-scale experiments while improving wall-clock efficiency. It offers a surprisingly simple yet effective path to scalable MoE training, preserving the benefits of large expert pools while reducing both training time and GPU cost.
Linghao Jin, Chufan Shi, Huijuan Wang +4
Cold Mixture-Of-Expert ModelsModern Mixture-Of-ExpertsExpert JudgmentSparsity
AskPDF