English
Related papers

Related papers: Training Foundation Models on a Full-Stack AMD Pla…

200 papers

We present ZAYA1-8B, a reasoning-focused mixture-of-experts (MoE) model with 700M active and 8B total parameters, built on Zyphra's MoE++ architecture. ZAYA1-8B's core pretraining, midtraining, and supervised fine-tuning (SFT) were…

Scaling Mixture-of-Experts (MoE) training introduces systems challenges absent in dense models. Because each token activates only a subset of experts, this sparsity allows total parameters to grow much faster than per-token computation,…

Emerging expert-specialized Mixture-of-Experts (MoE) architectures, such as DeepSeek-MoE, deliver strong model quality through fine-grained expert segmentation and large top-k routing. However, their scalability is limited by substantial…

Machine Learning · Computer Science 2025-08-20 Yueming Yuan , Ahan Gupta , Jianping Li , Sajal Dash , Feiyi Wang , Minjia Zhang

Mixture of Experts (MoE) models enhance neural network scalability by dynamically selecting relevant experts per input token, enabling larger model sizes while maintaining manageable computation costs. However, efficient training of…

Mixture-of-Experts (MoE) architectures have become standard in large language models, yet many of their core design choices - expert count, granularity, shared experts, load balancing, token dropping - have only been studied one or two at a…

Machine Learning · Computer Science 2026-05-13 Margaret Li , Sneha Kudugunta , Danielle Rothermel , Luke Zettlemoyer

Mixture-of-Experts (MoE) models can achieve promising results with outrageous large amount of parameters but constant computation cost, and thus it has become a trend in model scaling. Still it is a mystery how MoE layers bring quality…

Machine Learning · Computer Science 2021-08-10 An Yang , Junyang Lin , Rui Men , Chang Zhou , Le Jiang , Xianyan Jia , Ang Wang , Jie Zhang , Jiamang Wang , Yong Li , Di Zhang , Wei Lin , Lin Qu , Jingren Zhou , Hongxia Yang

The rapid scaling of large language models (LLMs) has unveiled critical limitations in current hardware architectures, including constraints in memory capacity, computational efficiency, and interconnection bandwidth. DeepSeek-V3, trained…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-24 Chenggang Zhao , Chengqi Deng , Chong Ruan , Damai Dai , Huazuo Gao , Jiashi Li , Liyue Zhang , Panpan Huang , Shangyan Zhou , Shirong Ma , Wenfeng Liang , Ying He , Yuqing Wang , Yuxuan Liu , Y. X. Wei

Training large-scale Mixture-of-Experts (MoE) models is bottlenecked by activation memory and expert-parallel communication, yet FP4 training remains impractical on Hopper-class GPUs without native MXFP4 or NVFP4 support. In this work, we…

Machine Learning · Computer Science 2026-03-04 Wuyue Zhang , Chongdong Huang , Chunbo You , Cheng Gu , Fengjuan Wang , Mou Sun

Large-scale Mixture of Experts (MoE) Large Language Models (LLMs) have recently become the frontier open-weight models, achieving remarkable model capability similar to proprietary ones. But their random expert selection mechanism…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-13 Zhongkai Yu , Yue Guan , Zihao Yu , Chenyang Zhou , Zhengding Hu , Shuyi Pei , Yangwook Kang , Yufei Ding , Po-An Tsai

Scaling model parameters improves model quality at the price of high computation overhead. Sparsely activated models, usually in the form of Mixture of Experts (MoE) architecture, have sub-linear scaling of computation cost with model size,…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-04-30 Jiamin Li , Yimin Jiang , Yibo Zhu , Cong Wang , Hong Xu

We present MegaScale-MoE, a production system tailored for the efficient training of large-scale mixture-of-experts (MoE) models. MoE emerges as a promising architecture to scale large language models (LLMs) to unprecedented sizes, thereby…

Training large-scale Mixture-of-Experts (MoE) models typically requires high-memory, high-bandwidth GPUs (e.g., A100), and their high cost has become a major barrier to large-model training. In contrast, affordable hardware is low-cost but…

Machine Learning · Computer Science 2026-01-13 Xin Ye , Daning Cheng , Boyang Zhang , Yunquan Zhang

Pretraining Large Language Models (LLMs) from scratch requires massive amount of compute. Aurora super computer is an ExaScale machine with 127,488 Intel PVC (Ponte Vechio) GPU tiles. In this work, we showcase LLM pretraining on Aurora at…

Machine Learning · Computer Science 2026-04-02 Dharma Teja Vooturi , Dhiraj Kalamkar , Dipankar Das , Bharat Kaul

The foundation-model ecosystem remains highly centralized because training requires immense compute resources and is therefore largely limited to large cloud operators. Edge-assisted foundation model training that harnesses spare compute on…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-04-14 Leyang Xue , Meghana Madhyastha , Myungjin Lee , Amos Storkey , Randal Burns , Mahesh K. Marina

We present MegaBlocks, a system for efficient Mixture-of-Experts (MoE) training on GPUs. Our system is motivated by the limitations of current frameworks, which restrict the dynamic routing in MoE layers to satisfy the constraints of…

Machine Learning · Computer Science 2022-11-30 Trevor Gale , Deepak Narayanan , Cliff Young , Matei Zaharia

Mixture-of-Experts (MoE) has emerged as a promising approach to scale up deep learning models due to its significant reduction in computational resources. However, the dynamic nature of MoE leads to load imbalance among experts, severely…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-01-16 Chenqi Zhao , Wenfei Wu , Linhai Song , Yuchen Xu , Yitao Yuan

Frontier models increasingly adopt Mixture-of-Experts (MoE) architectures to achieve large-model performance at reduced cost. However, training MoE models on HPC platforms is hindered by large memory footprints, frequent large-scale…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-07 Sajal Dash , Feiyi Wang

Training large language models (LLMs) efficiently requires a deep understanding of how modern GPU systems behave under real-world distributed training workloads. While prior work has focused primarily on kernel-level performance or…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-12-10 Marco Kurzynski , Shaizeen Aga , Di Wu

We present the design, implementation and engineering experience in building and deploying MegaScale, a production system for training large language models (LLMs) at the scale of more than 10,000 GPUs. Training LLMs at this scale brings…

‹ Prev 1 2 3 10 Next ›