FlashMoE: Reducing SSD I/O Bottlenecks via ML-Based Cache Replacement for Mixture-of-Experts Inference on Edge Devices

arXiv:2601.17063 · cs.LG, cs.AI · Submitted 2026-01-22 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-01-22

Updated: 2026-10-05

Code: https://github.com/ggml-org/llama.cpp

Terminology

• Edge Devices — 2× in this paper · explained in 2 episodes
• Mixture-of-Experts (MoE) — 1× in this paper · explained in 7 episodes
• Mixture-of-Experts (MoE) Models — 1× in this paper · explained in 3 episodes
• Sparsity — 1× in this paper · explained in 4 episodes

Sources

• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• Fast Inference of Mixture-of-Experts Language Models with Offloading
• Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
• Mixtral of Experts
• Scaling Laws for Neural Language Models
• OLMoE: Open Mixture-of-Experts Language Models
• Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
• HuggingFace's Transformers: State-of-the-art Natural Language Processing
• Qwen3 Technical Report

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks